{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import division, print_function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "from glob import glob\n",
    "import pandas\n",
    "import os\n",
    "import scipy.stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#os.environ[\"R_HOME\"] = \"d:/Schutte/R/R-3.5.0\"\n",
    "from rpy2.robjects.packages import importr\n",
    "from rpy2.robjects import pandas2ri\n",
    "pandas2ri.activate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "subjects = sorted([s[-2:] for s in glob(\"experiment_data/s_*\")])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "subject_demographics = pandas.DataFrame({\n",
    "    \"hz\": (30, \"f\"),\n",
    "    \"ls\": (25, \"f\"),\n",
    "    \"uo\": (22, \"m\"),\n",
    "    \"fn\": (20, \"f\"),\n",
    "    \"ql\": (21, \"f\"),\n",
    "    \"at\": (23, \"m\"),\n",
    "    \"cf\": (21, \"f\"),\n",
    "    \"sd\": (36, \"m\"),\n",
    "    \"qr\": (22, \"f\"),\n",
    "    \"op\": (22, \"f\"),\n",
    "    \"te\": (21, \"f\")}).T\n",
    "subject_demographics.columns = [\"age\", \"sex\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23.90909090909091"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subject_demographics.age.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "distance_data = {}\n",
    "contact_data = {}\n",
    "for subject in subjects:\n",
    "    dfs = []\n",
    "    for session in [1, 2]:\n",
    "        df = pandas.read_csv(os.path.join(\"experiment_data\", \"s_{}\".format(subject), \"ssn_%02d\" % session, \"results.csv\"))\n",
    "        df[\"subject\"] = subject\n",
    "        df[\"session\"] = session\n",
    "        dfs.append(df)\n",
    "    df = pandas.concat(dfs)\n",
    "    distance_data[subject] = df\n",
    "    \n",
    "    df = pandas.read_csv(os.path.join(\"experiment_data\", \"s_{}\".format(subject), \"ssn_03\", \"results.csv\"))\n",
    "    df[\"subject\"] = subject\n",
    "    contact_data[subject] = df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = next(iter(distance_data.values()))\n",
    "approach_times = unique(df[\"approach_time\"])\n",
    "distances = unique(df[\"distance\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = next(iter(contact_data.values()))\n",
    "start_distances = unique(df[\"start_distance\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Response histograms\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'dict' object has no attribute 'itervalues'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-19-ace90f4b6605>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpandas\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcontact_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitervalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m: 'dict' object has no attribute 'itervalues'"
     ]
    }
   ],
   "source": [
    "df = pandas.concat(contact_data.itervalues())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 11 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = subplots(3, 4, figsize=(4 * 3, 3 * 3))\n",
    "for i, ax in enumerate(axes.flat):\n",
    "    if i >= len(subjects):\n",
    "        ax.remove()\n",
    "        continue\n",
    "    subject = subjects[i]\n",
    "    const_resp, mod_resp = [df[\"response\"][(df[\"subject\"] == subject) & (df[\"condition\"] == c)]\n",
    "                            for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "    ax.hist([const_resp, mod_resp], range=(0, 8), bins=40, density=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([array([0.48349057, 0.40487421, 0.68003145, 0.76257862, 0.73113208,\n",
       "         0.71540881, 0.41666667, 0.17688679, 0.14544025, 0.07468553,\n",
       "         0.07861635, 0.05503145, 0.03930818, 0.05110063, 0.04716981,\n",
       "         0.03537736, 0.02751572, 0.01572327, 0.01572327, 0.01179245,\n",
       "         0.00393082, 0.00786164, 0.00393082, 0.00393082, 0.        ,\n",
       "         0.        , 0.00393082, 0.00393082, 0.        , 0.00393082,\n",
       "         0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "         0.        , 0.        , 0.        , 0.        , 0.        ]),\n",
       "  array([0.11006289, 0.22012579, 0.64072327, 0.63286164, 0.65644654,\n",
       "         0.98663522, 0.75471698, 0.22798742, 0.09433962, 0.07468553,\n",
       "         0.02358491, 0.        , 0.01572327, 0.01179245, 0.01572327,\n",
       "         0.02358491, 0.02358491, 0.09040881, 0.27515723, 0.02358491,\n",
       "         0.02751572, 0.01179245, 0.01179245, 0.01179245, 0.01572327,\n",
       "         0.00786164, 0.        , 0.        , 0.        , 0.00786164,\n",
       "         0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "         0.        , 0.        , 0.        , 0.00393082, 0.        ])],\n",
       " array([0. , 0.2, 0.4, 0.6, 0.8, 1. , 1.2, 1.4, 1.6, 1.8, 2. , 2.2, 2.4,\n",
       "        2.6, 2.8, 3. , 3.2, 3.4, 3.6, 3.8, 4. , 4.2, 4.4, 4.6, 4.8, 5. ,\n",
       "        5.2, 5.4, 5.6, 5.8, 6. , 6.2, 6.4, 6.6, 6.8, 7. , 7.2, 7.4, 7.6,\n",
       "        7.8, 8. ]),\n",
       " <a list of 2 Lists of Patches objects>)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT8AAAEyCAYAAACMONd1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAEBFJREFUeJzt3W+QXXddx/H3h4RaWiB1zOrUJLJ1DIwZxrG4E9DOYLWgKWUSH4AkDqgMEh9QBEGdoE6J9UkFx39jRTst8kdoJxTQDI0WR8qgjsVsW/6lpU4IgSxBuxQMImqpfn2wN52bzSZ7k5w9d3d/79fMTu859+T+fjtN3nvOuefcTVUhSa150rgnIEnjYPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKatHZcA69fv74mJyfHNbykVeq+++77SlVNLLbd2OI3OTnJ9PT0uIaXtEol+cIo23nYK6lJxk9Sk4yfpCYZP0lNWjR+Sd6e5JEknznD80nyR0kOJ/lUkud0P01J6tYoe37vALad5flrgc2Dr93A2y58WpK0tBaNX1V9DPjqWTbZAbyr5twLXJbk8q4mKElLoYtzfhuAY0PLM4N1p0myO8l0kunZ2dkOhpak89NF/LLAugV/MUhV3VJVU1U1NTGx6AXYkrRkuojfDLBpaHkjcLyD15WkJdNF/PYDPzt41/d5wImq+nIHrytJS2bRe3uT3A5cDaxPMgO8GXgyQFX9KXAAeBFwGPgm8MqlmuyysXfd0OMT45uHpPO2aPyqatcizxfwms5mJEk98A4PSU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWrSSPFLsi3Jw0kOJ9mzwPPfk+SeJA8k+VSSF3U/VUnqzqLxS7IGuBm4FtgC7EqyZd5mvwnsq6orgZ3An3Q9UUnq0ih7fluBw1V1pKoeA+4AdszbpoCnDx6vA453N0VJ6t4o8dsAHBtanhmsG7YXeHmSGeAA8NqFXijJ7iTTSaZnZ2fPY7qS1I1R4pcF1tW85V3AO6pqI/Ai4N1JTnvtqrqlqqaqampiYuLcZytJHRklfjPApqHljZx+WPsqYB9AVf0TcDGwvosJStJSGCV+B4HNSa5IchFzb2jsn7fNF4FrAJJ8P3Px87hW0rK1aPyq6nHgeuBu4CHm3tU9lOTGJNsHm70ReHWSTwK3Az9fVfMPjSVp2Vg7ykZVdYC5NzKG190w9PhB4KpupyZJS8c7PCQ1yfhJapLxk9Qk4yepSc3Gb3LPXUzuuWvc05A0Js3GT1LbjF/X9q6b+5K0rBk/SU0yfpKaZPwkNcn4SWqS8ZPUpJE+2KAlw9f+Hb3pujHORNJScs9PUpOMn6QmGT9JTfKc3zk4eT7w6MVjnoikC+aen6QmGT9JTTJ+kppk/CQ1yfhJapLxk9Qk4yepScZPUpNWffz8RUWSFrLq4ydJCzF+kppk/CQ1yfhJapLxk9Qk4yepScbvbPaum/uStOoYP0lNMn6SmmT8JDWpnd/hMXzubu+J8c1D0rLgnp+kJrWz53cm7hFKTXLPT1KTjJ+kJhk/SU0yfpKaNFL8kmxL8nCSw0n2nGGbn07yYJJDSd7b7TQlqVuLvtubZA1wM/BCYAY4mGR/VT04tM1m4E3AVVX1tSTfuVQTlqQujLLntxU4XFVHquox4A5gx7xtXg3cXFVfA6iqR7qdpiR1a5T4bQCODS3PDNYNeybwzCT/mOTeJNsWeqEku5NMJ5menZ09vxlLUgdGiV8WWFfzltcCm4GrgV3ArUkuO+0PVd1SVVNVNTUxMXGuc5WkzowSvxlg09DyRuD4Atv8VVV9q6o+DzzMXAwlaVkaJX4Hgc1JrkhyEbAT2D9vm78EfgwgyXrmDoOPdDlRSerSovGrqseB64G7gYeAfVV1KMmNSbYPNrsbeDTJg8A9wK9W1aNLNWlJulAjfbBBVR0ADsxbd8PQ4wLeMPiSpGXPOzwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNWlFxW9yz11M7rlr3NOQtAqsqPhJUleMn6QmGT9JTTJ+kppk/CQ1yfhJapLxk9Qk4yepScZPUpOMn6QmGT9JTTJ+kppk/CQ1yfhJapLxk9Qk4yepSWvHPYHVYPgDVo9ePMaJSBqZe36SmjRS/JJsS/JwksNJ9pxlu5ckqSRT3U1Rkrq3aPySrAFuBq4FtgC7kmxZYLunAb8EfLzrSUpS10Y557cVOFxVRwCS3AHsAB6ct91vA28BfqXTGY7glHNuN13X9/CSVqBRDns3AMeGlmcG656Q5EpgU1V96GwvlGR3kukk07Ozs+c8WUnqyijxywLr6oknkycBvw+8cbEXqqpbqmqqqqYmJiZGn6UkdWyU+M0Am4aWNwLHh5afBjwb+GiSo8DzgP2+6SFpORslfgeBzUmuSHIRsBPYf/LJqjpRVeurarKqJoF7ge1VNb0kM5akDiwav6p6HLgeuBt4CNhXVYeS3Jhk+1JPUJKWwkh3eFTVAeDAvHU3nGHbqy98WpK0tLzDQ1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGT1KTjJ+kJhk/SU1aO+4JqHF71w09PjG+eag57vlJapLxk9Qk4yepScZPUpOMn6QmGT9JTTJ+kppk/CQ1yfhJapLxk9Qk4yepScavL3vXnXofq6SxMn6SmrT64ucelqQRrL74SdIIRopfkm1JHk5yOMmeBZ5/Q5IHk3wqyd8leUb3U5Wk7iwavyRrgJuBa4EtwK4kW+Zt9gAwVVU/ANwJvKXriUpSl0bZ89sKHK6qI1X1GHAHsGN4g6q6p6q+OVi8F9jY7TQlqVujxG8DcGxoeWaw7kxeBfz1Qk8k2Z1kOsn07Ozs6LOUpI6NEr8ssK4W3DB5OTAFvHWh56vqlqqaqqqpiYmJ0WcpSR0b5RcYzQCbhpY3Asfnb5TkBcBvAD9aVf/TzfQkaWmMsud3ENic5IokFwE7gf3DGyS5EvgzYHtVPdL9NCWpW4vGr6oeB64H7gYeAvZV1aEkNybZPtjsrcBTgfcl+USS/Wd4OUlaFkb6vb1VdQA4MG/dDUOPX9DxvCRpSflLy5fQ5J67nnh89OIxTkTSaby9TVKTjJ+kJhk/SU0yfpKaZPwkNcn4SWqS8ZPUJOMnqUnGb0wm99x1ykXQkvpl/CQ1yfhJapLxk9Qk4yepScZPUpOMn6QmGT9JTTJ+kpq0Mj/Jee+6occnxjcPSSuWe36SmmT8JDXJ+Elq0so857eaeP5SGgv3/NSPvetODb00ZsZPUpOMn6Qmec5vhRj+4NOjN103xplIq4PxW2ZOidzFPzP3wDdCpM552CupScZPUpOMn6QmGT9JTTJ+q4C/BlM6d8avUQZTrfNSF53C6wnVCvf8JDXJPb9VzL046czc85PUJPf8NBL3IrXauOcnqUnGTxes98tm/GBUdcDD3pVo3B99f3L8+WN3Oa9xf49a9dzzk9Skkfb8kmwD/hBYA9xaVTfNe/7bgHcBPwQ8Crysqo52O1VdkDPtrXX9Z4ac+tmEi69fVtzzXPUWjV+SNcDNwAuBGeBgkv1V9eDQZq8CvlZV35dkJ/A7wMuWYsLq2DL7R34yjOcVxbN9L4sdqi+D7139GmXPbytwuKqOACS5A9gBDMdvB7B38PhO4I+TpKqqw7lKC7qgYJ6PPoK5zH4orUajxG8DcGxoeQZ47pm2qarHk5wAvgP4SheTlM7nULmPw+6zXf/4RJTP4brI3kM+TmMOfBbbOUvyUuAnq+oXBsuvALZW1WuHtjk02GZmsPy5wTaPznut3cDuweKzgIfPYa7rGW9MHb/d8Vv+3lfi+M+oqonFNhplz28G2DS0vBE4foZtZpKsBdYBX53/QlV1C3DLCGOeJsl0VU2dz5/tguO3O37L3/tqHn+US10OApuTXJHkImAnsH/eNvuBnxs8fgnwEc/3SVrOFt3zG5zDux64m7lLXd5eVYeS3AhMV9V+4Dbg3UkOM7fHt3MpJy1JF2qk6/yq6gBwYN66G4Ye/zfw0m6ndprzOlx2fMdf4WM7/hKNv+gbHpK0Gnl7m6QmGT9JTVoR8UuyLcnDSQ4n2dPz2G9P8kiSz/Q57mDsTUnuSfJQkkNJXtfz+Bcn+ecknxyM/1t9jj80jzVJHkjyoTGMfTTJp5N8Isn0GMa/LMmdST47+Hvwwz2O/azB933y6+tJXt/X+IM5/PLg795nktyepLPLv5f9Ob/BvcX/wtC9xcCuefcWL+X4zwe+Abyrqp7dx5hDY18OXF5V9yd5GnAf8FM9fu8BLq2qbyR5MvAPwOuq6t4+xh+axxuAKeDpVfXinsc+CkxV1Vgu8k3yTuDvq+rWwaVml1TVv49hHmuALwHPraov9DTmBub+zm2pqv9Ksg84UFXv6OL1V8Ke3xP3FlfVY8DJe4t7UVUfY4ELtnsa+8tVdf/g8X8ADzF3K2Ff41dVfWOw+OTBV68/LZNsBK4Dbu1z3OUgydOB5zN3KRlV9dg4wjdwDfC5vsI3ZC3wlMHNE5dw+g0W520lxG+he4t7C8BykWQSuBL4eM/jrknyCeAR4G+rqtfxgT8Afg34v57HPamADye5b3B7Zp++F5gF/nxw2H9rkkt7nsNJO4Hb+xywqr4E/C7wReDLwImq+nBXr78S4pcF1i3vY/WOJXkq8H7g9VX19T7Hrqr/raofZO62xq1Jejv0T/Ji4JGquq+vMRdwVVU9B7gWeM3gNEhf1gLPAd5WVVcC/wn0es4bYHC4vR14X8/jfjtzR3lXAN8NXJrk5V29/kqI3yj3Fq9ag3Nt7wfeU1UfGNc8BodbHwW29TjsVcD2wXm3O4AfT/IXPY5PVR0f/PcR4IPMnYbpywwwM7S3fSdzMezbtcD9VfVvPY/7AuDzVTVbVd8CPgD8SFcvvhLiN8q9xavS4A2H24CHqur3xjD+RJLLBo+fwtxfxs/2NX5VvamqNlbVJHP/3z9SVZ395F9MkksHbzQxONz8CaC3d/2r6l+BY0meNVh1Dad+jmZfdtHzIe/AF4HnJblk8G/hGubOe3di2f8CozPdW9zX+EluB64G1ieZAd5cVbf1NPxVwCuATw/OuwH8+uB2wz5cDrxz8E7fk4B9VdX75SZj9F3AB+f+3bEWeG9V/U3Pc3gt8J7BD/4jwCv7HDzJJcxdafGLfY4LUFUfT3IncD/wOPAAHd7qtuwvdZGkpbASDnslqXPGT1KTjJ+kJhk/SU0yfpKaZPwkNcn4SWrS/wOm4C/cSTLduwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = subplots(1, figsize=(5, 5))\n",
    "const_resp, mod_resp = [df[\"response\"][df[\"condition\"] == c] for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "ax.hist([const_resp, mod_resp], range=(0, 8), bins=40, density=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparion of medians\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pandas.concat(contact_data.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[[\"subject\", \"condition\", \"start_distance\", \"response\"]].to_csv(\"/tmp/quade.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum((df[\"start_distance\"] <= 4) & (df[\"condition\"] == \"rattle_mod\") & (df[\"response\"] == 0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['at', 'cf', 'fn', 'hz', 'ls', 'op', 'ql', 'qr', 'sd', 'te', 'uo']"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subjects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KstestResult(statistic=0.1804245283018868, pvalue=1.7931978737411935e-12)"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scipy.stats.kstest(df.query(\"condition == 'rattle_const' and start_distance <= 4\")[\"response\"],\n",
    "                   df.query(\"condition == 'rattle_mod' and start_distance <= 4\")[\"response\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.query(\"condition == 'rattle_const' and start_distance > 4\")[\"response\"].hist(bins=30)\n",
    "df.query(\"condition == 'rattle_mod' and start_distance > 4\")[\"response\"].hist(bins=30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "median_results = df.groupby([\"condition\", \"subject\", \"start_distance\"])[\"response\"].median()\n",
    "const_results = median_results.xs(\"rattle_const\")\n",
    "mod_results = median_results.xs(\"rattle_mod\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MannwhitneyuResult(statistic=184.0, pvalue=0.08855996466017407)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = np.array(const_results.reset_index()[\"start_distance\"] > 4)\n",
    "scipy.stats.mannwhitneyu(const_results[m], mod_results[m], alternative=\"less\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MannwhitneyuResult(statistic=701.0, pvalue=0.013073296417530897)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = np.array(const_results.reset_index()[\"start_distance\"] <= 4)\n",
    "scipy.stats.mannwhitneyu(const_results[m], mod_results[m], alternative=\"less\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ANOVA\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/michi/.py3/lib/python3.9/site-packages/rpy2/robjects/vectors.py:980: UserWarning: R object inheriting from \"POSIXct\" but without attribute \"tzone\".\n",
      "  warnings.warn('R object inheriting from \"POSIXct\" but without '\n",
      "R[write to console]: Warning:\n",
      "R[write to console]:  Converting \"subject\" to factor for ANOVA.\n",
      "\n",
      "R[write to console]: Warning:\n",
      "R[write to console]:  Converting \"condition\" to factor for ANOVA.\n",
      "\n",
      "R[write to console]: Warning:\n",
      "R[write to console]:  \"start_distance\" will be treated as numeric.\n",
      "\n",
      "R[write to console]: Warning:\n",
      "R[write to console]:  Collapsing data to cell means. *IF* the requested effects are a subset of the full design, you must use the \"within_full\" argument, else results may be inaccurate.\n",
      "\n",
      "R[write to console]: Warning:\n",
      "R[write to console]:  There is at least one numeric within variable, therefore aov() will be used for computation and no assumption checks will be obtained.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ez = importr(\"ez\")\n",
    "base = importr(\"base\")\n",
    "df = pandas.concat(contact_data.values())\n",
    "del df[\"trial_started\"]\n",
    "del df[\"trial_ended\"]\n",
    "aov = ez.ezANOVA(data=df[df[\"start_distance\"] > 4],\n",
    "                 dv=base.as_symbol(\"response\"),\n",
    "                 wid=base.as_symbol(\"subject\"),\n",
    "                 within=[base.as_symbol(\"condition\"), base.as_symbol(\"start_distance\")])\n",
    "                 #within=base.as_symbol(\"condition\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "subject  start_distance  condition   \n",
       "hz       1.414214        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         2.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         2.828427        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         4.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         5.656854        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         8.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "ls       1.414214        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         2.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         2.828427        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         4.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         5.656854        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "         8.000000        rattle_const    16\n",
       "                         rattle_mod      16\n",
       "Name: response, dtype: int64"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts = df.groupby([\"subject\", \"start_distance\", \"condition\"])[\"response\"].count()\n",
    "counts[counts != 20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "$ANOVA\n",
      "                    Effect DFn DFd         F          p p<.05         ges\n",
      "1                condition   1  10 8.8422020 0.01396000     * 0.348520801\n",
      "2           start_distance   1  10 7.0870921 0.02381087     * 0.149381880\n",
      "3 condition:start_distance   1  10 0.2387439 0.63565250       0.003501677\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(aov)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-subject, per-start-difference overshoot rates\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pandas.concat(contact_data.itervalues())\n",
    "df[\"overshoot\"] = (df[\"response\"] == 0)\n",
    "gr = df.groupby([\"subject\", \"start_distance\", \"condition\"])[\"overshoot\"]\n",
    "df = pandas.DataFrame(dict(n_overshoot=gr.sum(),\n",
    "                           n_no_overshoot=gr.count() - gr.sum(),\n",
    "                           n_total=gr.count()))[[\"n_overshoot\", \"n_no_overshoot\", \"n_total\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.41 0.000\n",
      "2.00 0.000\n",
      "2.83 0.140\n",
      "4.00 0.006\n",
      "5.66 0.030\n",
      "8.00 0.122\n"
     ]
    }
   ],
   "source": [
    "sum_df = df.groupby([\"start_distance\", \"condition\"])[[\"n_overshoot\", \"n_no_overshoot\"]].sum()\n",
    "for sd in start_distances:\n",
    "    print(\"{:.02f} {:.03f}\".format(sd, scipy.stats.fisher_exact(sum_df.loc[sd])[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "at 1.41 0.047\n",
      "at 2.00 1.000\n",
      "at 2.83 1.000\n",
      "at 4.00 1.000\n",
      "at 5.66 1.000\n",
      "at 8.00 1.000\n",
      "cf 1.41 0.047\n",
      "cf 2.00 0.000\n",
      "cf 2.83 0.047\n",
      "cf 4.00 0.106\n",
      "cf 5.66 1.000\n",
      "cf 8.00 0.106\n",
      "fn 1.41 0.235\n",
      "fn 2.00 0.342\n",
      "fn 2.83 1.000\n",
      "fn 4.00 0.487\n",
      "fn 5.66 1.000\n",
      "fn 8.00 1.000\n",
      "hz 1.41 0.101\n",
      "hz 2.00 0.484\n",
      "hz 2.83 0.333\n",
      "hz 4.00 0.226\n",
      "hz 5.66 0.484\n",
      "hz 8.00 1.000\n",
      "ls 1.41 0.226\n",
      "ls 2.00 1.000\n",
      "ls 2.83 1.000\n",
      "ls 4.00 1.000\n",
      "ls 5.66 1.000\n",
      "ls 8.00 1.000\n",
      "op 1.41 0.333\n",
      "op 2.00 0.273\n",
      "op 2.83 0.091\n",
      "op 4.00 1.000\n",
      "op 5.66 0.487\n",
      "op 8.00 1.000\n",
      "ql 1.41 1.000\n",
      "ql 2.00 1.000\n",
      "ql 2.83 1.000\n",
      "ql 4.00 1.000\n",
      "ql 5.66 1.000\n",
      "ql 8.00 1.000\n",
      "qr 1.41 1.000\n",
      "qr 2.00 1.000\n",
      "qr 2.83 1.000\n",
      "qr 4.00 1.000\n",
      "qr 5.66 1.000\n",
      "qr 8.00 1.000\n",
      "sd 1.41 0.231\n",
      "sd 2.00 1.000\n",
      "sd 2.83 1.000\n",
      "sd 4.00 1.000\n",
      "sd 5.66 1.000\n",
      "sd 8.00 1.000\n",
      "te 1.41 1.000\n",
      "te 2.00 1.000\n",
      "te 2.83 1.000\n",
      "te 4.00 1.000\n",
      "te 5.66 1.000\n",
      "te 8.00 1.000\n",
      "uo 1.41 0.003\n",
      "uo 2.00 0.003\n",
      "uo 2.83 0.182\n",
      "uo 4.00 0.487\n",
      "uo 5.66 1.000\n",
      "uo 8.00 1.000\n"
     ]
    }
   ],
   "source": [
    "for subject in subjects:\n",
    "    for sd in start_distances:\n",
    "        print(\"{} {:.02f} {:.03f}\".format(subject, sd, scipy.stats.fisher_exact(df[[\"n_overshoot\", \"n_no_overshoot\"]].loc[subject, sd])[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-subject, per-start-difference scatter plots\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 11 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pandas.concat(contact_data.itervalues())\n",
    "fig, axes = subplots(3, 4, figsize=(4 * 3, 3 * 3))\n",
    "for i, ax in enumerate(axes.flat):\n",
    "    if i >= len(subjects):\n",
    "        ax.remove()\n",
    "        continue\n",
    "    subject = subjects[i]\n",
    "    sub_df = df[df[\"subject\"] == subject]\n",
    "    for cond in [\"rattle_const\", \"rattle_mod\"]:\n",
    "        sub_df = df[(df[\"subject\"] == subject) & (df[\"condition\"] == cond)]\n",
    "        ax.scatter(sub_df[\"start_distance\"] + numpy.random.uniform(-0.15, 0.15, size=len(sub_df)), sub_df[\"response\"], s=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-subject, per-start-distance difference pairs\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "result_dfs = []\n",
    "\n",
    "for subject in subjects:\n",
    "    df = contact_data[subject]\n",
    "    for start_distance in start_distances:\n",
    "        sub_df = df[df[\"start_distance\"] == start_distance]\n",
    "        const_df, mod_df = [sub_df[sub_df[\"condition\"] == c] for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "        const_responses = array(const_df[\"response\"])\n",
    "        mod_responses = array(mod_df[\"response\"])\n",
    "        differences = (mod_responses[newaxis, :] - const_responses[:, newaxis]).flatten()\n",
    "        result_dfs.append(pandas.DataFrame(dict(subject=subject, start_distance=start_distance,\n",
    "                                                mod_const_difference=differences)))\n",
    "\n",
    "diff_df = pandas.concat(result_dfs)[[\"subject\", \"start_distance\", \"mod_const_difference\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hist(diff_df[\"mod_const_difference\"], bins=50);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 11 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = subplots(3, 4, figsize=(4 * 3, 3 * 3))\n",
    "for i, ax in enumerate(axes.flat):\n",
    "    if i >= len(subjects):\n",
    "        ax.remove()\n",
    "        continue\n",
    "    subject = subjects[i]\n",
    "    df = diff_df[diff_df[\"subject\"] == subject]\n",
    "    ax.hist(df[\"mod_const_difference\"], range=(-5, 5), bins=50, density=True)\n",
    "    ax.vlines([0, df[\"mod_const_difference\"].mean()], 0, 1.2, lw=1, linestyles=[\"dashed\", \"solid\"])\n",
    "    ax.set_ylim(0, 1.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x2376 with 66 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = subplots(len(subjects), len(start_distances), figsize=(len(start_distances) * 3, len(subjects) * 3))\n",
    "for row, subject in zip(axes, subjects):\n",
    "    for ax, start_distance in zip(row, start_distances):\n",
    "        df = diff_df[(diff_df[\"subject\"] == subject) & (diff_df[\"start_distance\"] == start_distance)]\n",
    "        ax.hist(df[\"mod_const_difference\"], range=(-5, 5), bins=50, density=True)\n",
    "        ax.set_ylim(0, 1.5)\n",
    "        ax.text(-5, 0.1, \"{:.02f}\".format(df[\"mod_const_difference\"].mean()))\n",
    "        if scipy.stats.wilcoxon(df[\"mod_const_difference\"]).pvalue < 0.05:\n",
    "            ax.text(-5, 0.2, \"*\", fontdict=dict(fontsize=16))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-subject, per-start-distance mean differences\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "gr = pandas.concat(contact_data.itervalues()).groupby([\"condition\", \"start_distance\", \"subject\"])[\"response\"]\n",
    "mean_responses = gr.mean()\n",
    "mean_diff = array(mean_responses[\"rattle_mod\"]) - array(mean_responses[\"rattle_const\"])\n",
    "mean_diff_df = pandas.DataFrame(index=mean_responses[\"rattle_mod\"].index)\n",
    "mean_diff_df[\"const_response\"] = array(mean_responses[\"rattle_const\"])\n",
    "mean_diff_df[\"mod_response\"] = array(mean_responses[\"rattle_mod\"])\n",
    "mean_diff_df[\"response_diff\"] = mean_diff_df[\"mod_response\"] - mean_diff_df[\"const_response\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test differences between const and mod stopping distances for normality."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p (H₀: responses are normally distributed) = 8.80486459209e-06\n"
     ]
    }
   ],
   "source": [
    "print(\"p (H₀: responses are normally distributed) =\", scipy.stats.shapiro(mean_diff_df[\"response_diff\"])[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "They obviously do not follow a normal distribution.  Thus, test whole dataset and individual stopping distances for const vs. mod differences using a nonparametric test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "H₀: mod and const responses don't differ\n",
      "p (globally) = 2.79e-08\n",
      "p (start distance 1.41 m) = 0.033\n",
      "p (start distance 2.00 m) = 0.021\n",
      "p (start distance 2.83 m) = 0.013\n",
      "p (start distance 4.00 m) = 0.016\n",
      "p (start distance 5.66 m) = 0.016\n",
      "p (start distance 8.00 m) = 0.050\n"
     ]
    }
   ],
   "source": [
    "print(\"H₀: mod and const responses don't differ\")\n",
    "print(\"p (globally) = {:.3g}\".format(scipy.stats.wilcoxon(mean_diff_df[\"response_diff\"]).pvalue))\n",
    "for start_distance in start_distances:\n",
    "    print(\"p (start distance {:.2f} m) = {:.3f}\".format(\n",
    "        start_distance,\n",
    "        scipy.stats.wilcoxon(mean_diff_df[\"response_diff\"].loc[start_distance]).pvalue))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Permutation tests\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def permute(df, random_state):\n",
    "    indices = []\n",
    "    for subject in subjects:\n",
    "        for start_distance in start_distances:\n",
    "            cond_indices = np.where((df[\"subject\"] == subject) & (df[\"start_distance\"] == start_distance))[0]\n",
    "            random_state.shuffle(cond_indices)\n",
    "            indices.extend(cond_indices)\n",
    "    return df.iloc[indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pandas.concat(contact_data.itervalues()).sort_values([\"condition\", \"subject\", \"start_distance\"])\n",
    "const_df = df[df[\"condition\"] == \"rattle_const\"]\n",
    "mod_df = df[df[\"condition\"] == \"rattle_mod\"]\n",
    "shuffled_diff_dfs = []\n",
    "rs = np.random.RandomState(54321)\n",
    "\n",
    "for i in range(10):\n",
    "    shuffled_mod_df = permute(mod_df, rs)\n",
    "    df = pandas.DataFrame()\n",
    "    df[\"subject\"] = array(const_df[\"subject\"])\n",
    "    df[\"start_distance\"] = array(const_df[\"start_distance\"])\n",
    "    df[\"const_response\"] = array(const_df[\"response\"])\n",
    "    df[\"mod_response\"] = array(shuffled_mod_df[\"response\"])\n",
    "    df[\"response_diff\"] = array(shuffled_mod_df[\"response\"]) - array(const_df[\"response\"])\n",
    "    shuffled_diff_dfs.append(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Contigency table test on overshoot information\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "barnard = importr(\"Barnard\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pandas.concat(contact_data.itervalues())\n",
    "df[\"overshot\"] = (df[\"response\"] == 0)\n",
    "df[\"not_overshot\"] = (df[\"response\"] != 0)\n",
    "df[\"trial_type\"] = where(df[\"start_distance\"] <= 4, \"short\", \"long\")\n",
    "df = df.groupby([\"trial_type\", \"condition\"])[[\"overshot\", \"not_overshot\"]].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>overshot</th>\n",
       "      <th>not_overshot</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>condition</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>rattle_const</th>\n",
       "      <td>110.0</td>\n",
       "      <td>738.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rattle_mod</th>\n",
       "      <td>26.0</td>\n",
       "      <td>822.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              overshot  not_overshot\n",
       "condition                           \n",
       "rattle_const     110.0         738.0\n",
       "rattle_mod        26.0         822.0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[\"short\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.030660377358490566"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "26/(26+822)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "u'the label [5.65685424949] is not in the [index]'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0mTraceback (most recent call last)",
      "\u001b[1;32m<ipython-input-33-4dce06748adc>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mscipy\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mstats\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfisher_exact\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mstart_distances\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0malternative\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"greater\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mc:\\python27\\lib\\site-packages\\pandas\\core\\indexing.pyc\u001b[0m in \u001b[0;36m__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   1371\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1372\u001b[0m             \u001b[0mmaybe_callable\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcom\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_apply_if_callable\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mobj\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1373\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_getitem_axis\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmaybe_callable\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1374\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1375\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_is_scalar_access\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\python27\\lib\\site-packages\\pandas\\core\\indexing.pyc\u001b[0m in \u001b[0;36m_getitem_axis\u001b[1;34m(self, key, axis)\u001b[0m\n\u001b[0;32m   1624\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1625\u001b[0m         \u001b[1;31m# fall thru to straight lookup\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1626\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_has_valid_type\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1627\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_label\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1628\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\python27\\lib\\site-packages\\pandas\\core\\indexing.pyc\u001b[0m in \u001b[0;36m_has_valid_type\u001b[1;34m(self, key, axis)\u001b[0m\n\u001b[0;32m   1512\u001b[0m                 \u001b[1;32mraise\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1513\u001b[0m             \u001b[1;32mexcept\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1514\u001b[1;33m                 \u001b[0merror\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1515\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1516\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mTrue\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\python27\\lib\\site-packages\\pandas\\core\\indexing.pyc\u001b[0m in \u001b[0;36merror\u001b[1;34m()\u001b[0m\n\u001b[0;32m   1499\u001b[0m                 raise KeyError(u\"the label [{key}] is not in the [{axis}]\"\n\u001b[0;32m   1500\u001b[0m                                .format(key=key,\n\u001b[1;32m-> 1501\u001b[1;33m                                        axis=self.obj._get_axis_name(axis)))\n\u001b[0m\u001b[0;32m   1502\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1503\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: u'the label [5.65685424949] is not in the [index]'"
     ]
    }
   ],
   "source": [
    "scipy.stats.fisher_exact(df.loc[start_distances[4]].values, alternative=\"greater\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main plot\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "matplotlib.rcParams.update({\"font.family\": \"Arial\", \"font.size\": 8,\n",
    "                            \"xtick.major.width\": 0.5, \"ytick.major.width\": 0.5,\n",
    "                            \"xtick.minor.width\": 0.5, \"ytick.minor.width\": 0.5,\n",
    "                            \"axes.linewidth\": 0.5})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bracket(ax, pos=[0,0], scalex=1, scaley=1, text=\"\",textkw = {}, linekw = {}):\n",
    "    x = np.array([0, 0.05, 0.45,0.5])\n",
    "    y = np.array([0,-0.01,-0.01,-0.02])\n",
    "    x = np.concatenate((x,x+0.5)) \n",
    "    y = np.concatenate((y,y[::-1]))\n",
    "    ax.plot(x*scalex+pos[0], y*scaley+pos[1], clip_on=False, **linekw)\n",
    "    ax.text(pos[0]+0.5*scalex, (y.min()-0.02)*scaley+pos[1], text, ha=\"center\", va=\"center\", **textkw)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_histograms(fig, ax, overshoot_ax, query, legend_ax):\n",
    "    df = pandas.concat(contact_data.itervalues()).query(query)\n",
    "    nonzero_mask = (df[\"response\"] > 0)\n",
    "    df[\"overshot\"] = ~nonzero_mask\n",
    "    df[\"not_overshot\"] = nonzero_mask\n",
    "    overshot_table = df.groupby(\"condition\")[[\"overshot\", \"not_overshot\"]].sum().values\n",
    "    const_resp, mod_resp = [-df[\"response\"][(df[\"condition\"] == c) & nonzero_mask] for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "    #_, bins, _ = ax.hist(const_resp, range=(-8, 0), bins=40, edgecolor=\"k\", lw=0.5, alpha=0.7, density=True)\n",
    "    #ax.hist(mod_resp, bins=bins, alpha=0.5, edgecolor=\"k\", lw=0.5, density=True)\n",
    "    _, bins, patches = ax.hist([const_resp, mod_resp], range=(-8, 0), bins=40, density=True)\n",
    "    print(median(const_resp), median(mod_resp))\n",
    "    \n",
    "    if scipy.stats.mannwhitneyu(const_resp, mod_resp, alternative=\"greater\")[1] < 0.05:\n",
    "        ax.text(-1, 0.2 / (bins[1] - bins[0]), \"*\", fontdict=dict(fontsize=12), verticalalignment=\"center\")\n",
    "\n",
    "    const_overshoot = (~nonzero_mask & (df[\"condition\"] == \"rattle_const\")).sum()\n",
    "    const_total = (df[\"condition\"] == \"rattle_const\").sum()\n",
    "    const_height = const_overshoot / const_total / (bins[1] - bins[0])\n",
    "    mod_overshoot = (~nonzero_mask & (df[\"condition\"] == \"rattle_mod\")).sum()\n",
    "    mod_total = (df[\"condition\"] == \"rattle_mod\").sum()\n",
    "    mod_height = mod_overshoot / mod_total / (bins[1] - bins[0])\n",
    "    overshoot_ax.bar([-1, 0], [const_height, mod_height],\n",
    "                     align=\"edge\", color=[patches[0][0].get_facecolor(), patches[1][0].get_facecolor()])\n",
    "\n",
    "    const_ci_lo, _ = statsmodels.stats.proportion.proportion_confint(const_overshoot, const_total)\n",
    "    const_ci_height = const_height - const_ci_lo / (bins[1] - bins[0])\n",
    "    mod_ci_lo, _ = statsmodels.stats.proportion.proportion_confint(mod_overshoot, mod_total)\n",
    "    mod_ci_height = mod_height - mod_ci_lo / (bins[1] - bins[0])\n",
    "    overshoot_ax.errorbar([-0.6, 0.4], [const_height, mod_height],\n",
    "                          [const_ci_height, mod_ci_height],\n",
    "                          color=\"k\", lw=0.7, linestyle=\"none\")\n",
    "    print(const_overshoot / const_total, const_overshoot / const_total - const_ci_lo)\n",
    "    print(mod_overshoot / mod_total, mod_overshoot / mod_total - mod_ci_lo)\n",
    "\n",
    "    print(scipy.stats.fisher_exact(overshot_table, alternative=\"greater\")[1])\n",
    "    if scipy.stats.fisher_exact(overshot_table, alternative=\"greater\")[1] < 0.05:\n",
    "        overshoot_ax.text(0, max(const_height, mod_height) + 0.15, \"*\", fontdict=dict(fontsize=12),\n",
    "                          verticalalignment=\"center\", horizontalalignment=\"center\")\n",
    "    overshoot_ax.set_xlim(-1.5, 1.5)\n",
    "    overshoot_ax.set_xticks([0])\n",
    "    overshoot_ax.set_xticklabels([\"miss\"])\n",
    "    overshoot_ax.set_yticklabels([])\n",
    "\n",
    "    ax.set_xlim(-8, 0)\n",
    "    ax.set_xticks([-7, -5, -3, -1])\n",
    "    ax.set_xticks([-6, -4, -2, 0], minor=True)\n",
    "    ax.set_ylabel(\"% of trials stopped\\nat this distance\")\n",
    "    for item in [ax, overshoot_ax]:\n",
    "        item.spines[\"right\"].set_visible(False)\n",
    "        item.spines[\"top\"].set_visible(False)\n",
    "        item.set_ylim(0, 0.23 / (bins[1] - bins[0]))\n",
    "        item.yaxis.set_major_locator(ticker.MultipleLocator(0.05 / (bins[1] - bins[0])))\n",
    "    ax.xaxis.set_major_formatter(ticker.FuncFormatter(lambda x, pos: \"%d\" % -x))\n",
    "    ax.yaxis.set_major_formatter(ticker.PercentFormatter(xmax=1 / (bins[1] - bins[0]), decimals=0, symbol=\"\"))\n",
    "    if legend_ax:\n",
    "        legend_ax.legend(patches,\n",
    "                         [\"constant\\nRatFreq\", \"modulated\\nRatFreq\"],\n",
    "                         fontsize=\"small\", loc=(0.7, 0.5))\n",
    "    ax.set_axisbelow(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_modulation_function(fig, ax):\n",
    "    x = linspace(0, 10, 100)\n",
    "    static = 12 * ones(len(x))\n",
    "    modulated = 70 * ones(len(x))\n",
    "    modulated[x > 4] = 35 - 3.75 * x[x > 4]\n",
    "    modulated[x > 8] = 0\n",
    "    h0, = ax.plot(x, static, c=\"C0\", lw=1)\n",
    "    h1, = ax.plot(x[x < 4], modulated[x < 4], c=\"C1\", lw=1)\n",
    "    ax.plot(x[(x >= 4) & (x < 8)], modulated[(x >= 4) & (x < 8)], c=\"C1\", lw=1)\n",
    "    ax.plot(x[x >= 8], modulated[x >= 8], c=\"C1\", lw=1)\n",
    "    \n",
    "    for sd in start_distances:\n",
    "        h = ax.arrow(sd, 45, -0.25, 0, head_width=4, head_length=0.15, color=\"k\", lw=0.5)\n",
    "        ax.scatter([sd], [45], marker=\"o\", c=\"k\", s=2)\n",
    "    ax.text((start_distances[-1] + start_distances[-2] - 0.4) / 2, 40,\n",
    "            \"start\\npositions\", horizontalalignment=\"center\", verticalalignment=\"center\",\n",
    "            fontsize=\"small\")\n",
    "\n",
    "    ax.set_ylim(0, 80)\n",
    "    ax.set_yticks([5, 12, 20, 70])\n",
    "    lbl = ax.yaxis.get_majorticklabels()[0]\n",
    "    lbl.set_transform(lbl.get_transform() + matplotlib.transforms.ScaledTranslation(0, -0.05, fig.dpi_scale_trans))\n",
    "    lbl = ax.yaxis.get_majorticklabels()[2]\n",
    "    lbl.set_transform(lbl.get_transform() + matplotlib.transforms.ScaledTranslation(0, 0.05, fig.dpi_scale_trans))\n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "    ax.set_xlim(8, 0)\n",
    "    ax.set_xticks([7, 5, 3, 1])\n",
    "    ax.set_xticks([8, 6, 4, 2, 0], minor=True)\n",
    "    #ax.set_xticklabels([])\n",
    "    ax.set_xlabel(\"Distance from snake (m)\", labelpad=1)\n",
    "    ax.set_ylabel(\"RatFreq (Hz)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_overshooting(fig, ax):\n",
    "    df = pandas.concat(contact_data.itervalues())\n",
    "    df[\"trial_type\"] = np.where(df[\"start_distance\"] > 4, \"long\", \"short\")\n",
    "    df[\"overshot\"] = (df[\"response\"] == 0)\n",
    "    df[\"not_overshot\"] = (df[\"response\"] != 0)\n",
    "    gr = df.groupby([\"trial_type\", \"condition\"])[[\"overshot\", \"not_overshot\"]]\n",
    "    df = gr.sum()\n",
    "    long_star = scipy.stats.fisher_exact(df.loc[\"long\"].values, alternative=\"greater\")[1] < 0.05\n",
    "    short_star = scipy.stats.fisher_exact(df.loc[\"short\"].values, alternative=\"greater\")[1] < 0.05\n",
    "\n",
    "    df = pandas.concat(contact_data.itervalues())\n",
    "    df[\"overshot\"] = (df[\"response\"] == 0)\n",
    "    df[\"start_distance\"] = -df[\"start_distance\"]\n",
    "    gr = df.groupby([\"condition\", \"start_distance\"])[\"overshot\"]\n",
    "    df = gr.sum() / gr.count()\n",
    "    \n",
    "    for i, condition in enumerate([\"rattle_const\", \"rattle_mod\"]):\n",
    "        ax.bar(arange(len(start_distances)) - 0.15 + 0.3 * i, df.loc[condition].values, 0.3,\n",
    "               edgecolor=\"k\", lw=0.7, alpha=0.7)\n",
    "\n",
    "    if long_star:\n",
    "        bracket(ax, [0-0.3, 0.04], scalex=2-0.3, scaley=-1, text=\"*\",\n",
    "                textkw=dict(fontdict=dict(fontsize=12)),\n",
    "                linekw=dict(color=\"k\"))\n",
    "    if short_star:\n",
    "        bracket(ax, [2-0.3, 0.21], scalex=4-0.3, scaley=-1, text=\"*\",\n",
    "                textkw=dict(fontdict=dict(fontsize=12)),\n",
    "                linekw=dict(color=\"k\"))\n",
    "\n",
    "    ax.set_xticks(arange(len(start_distances)))\n",
    "    ax.set_xticklabels([\"{:.1f}\".format(sd) for sd in start_distances[::-1]])\n",
    "    ax.set_ylim(0, 0.28)\n",
    "    ax.set_yticks(np.linspace(0, 0.2, 3))\n",
    "    ax.yaxis.set_major_formatter(ticker.PercentFormatter(xmax=1, decimals=0))\n",
    "    ax.set_axisbelow(True)\n",
    "\n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "    ax.set_title(\"Frequency of overshooting\", fontsize=\"medium\")\n",
    "    ax.set_xlabel(\"Starting distance [m]\")\n",
    "    ax.set_ylabel(\"Perc. of trials not\\nstopped in time\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_acoustic_stimulus(wave_ax, spect_ax):\n",
    "    fs, signal = scipy.io.wavfile.read(\"stimulus.wav\")\n",
    "    signal = signal[int(0.95 * fs):int(2.45 * fs)]\n",
    "    t = arange(len(signal)) / fs\n",
    "    wave_ax.plot(t, signal, c=\"C1\", lw=0.5)\n",
    "    wave_ax.set_ylim(wave_ax.get_ylim()[0] * 1.2, wave_ax.get_ylim()[1] * 1.6)\n",
    "    wave_ax.xaxis.set_major_locator(MultipleLocator(0.5))\n",
    "    wave_ax.xaxis.set_minor_locator(MultipleLocator(0.1))\n",
    "    \n",
    "    _, _, _, img = spect_ax.specgram(signal + numpy.random.normal(scale=3500), NFFT=256, Fs=44100, noverlap=250,\n",
    "                                     cmap=\"jet\")\n",
    "    # full range: -144, 35\n",
    "    img.set_clim(-50, 35)\n",
    "    spect_ax.set_ylim(0, 10000)\n",
    "    spect_ax.set_ylabel(\"Freq. (kHz)\")\n",
    "    spect_ax.yaxis.set_major_locator(MultipleLocator(5000))\n",
    "    spect_ax.yaxis.set_major_formatter(FuncFormatter(lambda x, loc: \"{:.0f}\".format(x / 1000)))\n",
    "    spect_ax.spines[\"bottom\"].set_visible(False)\n",
    "    spect_ax.spines[\"left\"].set_visible(False)\n",
    "    spect_ax.set_xticks([])\n",
    "\n",
    "    for ax in [wave_ax, spect_ax]:\n",
    "        ax.set_xlim(0, 1.5)\n",
    "        ax.spines[\"right\"].set_visible(False)\n",
    "        ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "    wave_ax.yaxis.set_visible(False)\n",
    "    wave_ax.spines[\"left\"].set_visible(False)\n",
    "    wave_ax.set_xlabel(\"Stimulus time (s)\")\n",
    "    spect_ax.set_xticklabels([])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "def label_frame(fig, overall_ax, ax, xshift, yshift, letter):\n",
    "    w, h = fig.bbox_inches.width, fig.bbox_inches.height\n",
    "    bbox = ax.get_position() * array([w, h])\n",
    "    x = bbox[0, 0] + xshift\n",
    "    y = bbox[1, 1] + yshift\n",
    "    overall_ax.text(x / w, y / h, letter, fontdict=dict(fontsize=12))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.8294807374477384 -0.9849396049976351\n",
      "0.12971698113207547 0.022614075413668486\n",
      "0.030660377358490566 0.011603176416911068\n",
      "8.991085004716432e-15\n",
      "-1.078594982624055 -1.21456921100616\n",
      "0.02830188679245283 0.015784780140662858\n",
      "0.0023584905660377358 0.0023584905660377358\n",
      "0.001598754746178463\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAASIAAAE3CAYAAAAOpUyeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsnXd4lMXWwH+HBBICIZESYigGAUG6AkYUBREFFUVsKFcFr1zAa/dawJpYUbBeewPLh+1aUewKKk0Rg4KA9B6IlJCEBFLO98e8m60hIWRLyPyeZ5/dPTPzznl3syczZ86cEVXFYrFYwkmdcCtgsVgs1hBZLJawYw2RxWIJO9YQWSyWsGMNkcViCTvWEFkslrBjDZHFYgk71hBZLJawYw2RxWIJO9YQWSyWsGMNkcViCTvWEFkslrBjDZHFYgk71hBZIhoR6S8ii8OthyW4WENksVjCTsgMkYjUEZEnRWS+iPwpIktF5MRQ9W+p0TQUkf+JSKaIzBSRo8KtkKV6CeWIKA1IAfqoaifgNWB8CPu31FxaAY+pag9gGvBGmPWxVDMSygyNItIBGAC0BfoDuap6SsgUsNQ4RKQ/8LiqHuO8rwcUAI1VNSeculmqj1BOzc4CPnPefgw8D0io+rfUaEp83itQFA5FLMEhlFOz04DpqvocsAA4F4gKYf+Wmkt3EenhvB4L/KSqe8KpkKV6CaUheh7oLyJ/AAuBVUAbEbErd5aKWArcIyKLgHOAkWHWx1LNhNRHZLFYLIGIDrcC+0NEWgODgVLgC1XdWEH9dsB7qnqMiNwCpAIJwI2qmh1sfS0WS9WIWEMkItdjlvvnYxyT14rISlV9qZz6ycBoIF9EYoGTVfVsETkF+BfwoE/9/piVO2DkPZCKTsvg40tO59xWX8LGdK/rr9IXOJHZZMl7mEUbN/puBssuPIKj49dCnrudbs2AxXD2gHf5VJZ4t3k+g05jf2WpfOIln6ClvMxosuVV7/q/ZPBeryFcJJ8AGUb2Wwa7u9Yloe8+mOfR79MZfHr1AM4+/FvIMnJdnQGZICkKxxvZ+dqO/227jO+S+nCq3IX5qEH/yKBnlx9ZKN+4rzktgx6XzGWRfOGWDcggccYWcmKnAbvRTRkwFaStwsVOv3dnwCWQ1HFd2T3p8xkUXQz1Ril8lI5qepUWLUaPHq0tW7asSlPLQZKRkfGKqo6urutFrCECPlLVdR7vp4tIanmVVTULGC8iXwCNgW1O0UaMQfOtPxOYCSCSfg8ArSGRXXApMNG7/pGZWQzq8SWvNbzNy9gAUAx3cr+/fBsQBdtp4q9wD4iiBGgObC0T38wkpjDKv/4KKO4VBT0EMh3ZMzDlpStgHDDPo24apLAFbsAdqfUo0A+S01aT5YiiKIE1QBJw/BkwzxiiV7tcwsLP+kLLvmUGOeuSBFLYzCJPnXpB25iVLORoYD7sAE73uh1YYT6f7E1JbtlP8MLYK6Gj/20eCC1btiQ9Pf3gLmKpEhkZGfudnRwoEesoVtV1IvKuiDQAEJEHVHVtJZtvg7Jff0tgc2VbzeEE+DRA2WJoynYoDlC2Bh5iArRL95YnAbEw99cB/m2ehLG8ABzrJW78XiFxPiMuAJrDBlpBrIcsHy7gf9DQp+490Cl/qfe/mc2mfgz7ykT12Ie2h82keF2jE3/S96yvjQl3SF6Rw+e/nufdT1c4kxnOjTp9pABTPeosBWZD3Vh3v+yAsTmv+N+jpdYSsYbIQYA3RSQF75/gflHVYuB7EXkWGAM8U6mGS+EkfjBeKR/yL6xDKzYYr5MPC28/mj/oChd7y59OuhK2Qt3U3UBd78IV8BTXUvYjdnE8pLLGv5MkyCUednnIFsApfGdMrSePwv8anO/9iZ1o7m/X3sQyUTtWsrVxAss5Cv72UGHBIn5aNRATruMwA4b2fMu7nzbwK72A6eb9UmAHHPHesrIq+fPqQB50aLLc694LGtaFHlgsQOQbopXAVcCLQPvKNFDVwc7zk6r6b1W9pNIRuCtgKZ1MhJMPDWaUmv/+/f3Ljn1kKefN+Jwj7/P2A43OfwV2wzNNrgbqezcaDa8zEj8rcg6cwBz/TubBMD6CO92iP1fAgpLe/hPscXDph+/DYA9D8iuwF1Jj1paJ2rKK5EdyeCL/RjMddXEH5LWMhvs9XDf5cIe3mw0ehp4sAAaa97HANbCXemVVGlxTCplwEj+627WBRj8V0fiCTf73aamVRLohmuv4fkYAuUHvLQGS2AZfBChbCrtI9PbFuPgElp15BKu/7Owljn3etGvCdmC3d5utkMRWfB3fvAUzONO/j0dhFW1hrVvUPAoafVJE4x4+P+heQCEkt/UYWa0HcjCjOod4cqE9LGrQzfu+boc9MXFeBi7j9lvZ6jt6S4HtNKVsvno40BhOMa43Q7552ubZ9kSgOexY2cL/Pi21kog1RCJyG9BRRG7FuGMXVdDk4PkEht78ld8UC0AfhR45iwNu092TCR1nrCN50Gov+e47gHmQylp8fUHb73V8M+0aeV9sFOwjxr+TWMe57OHgbdIPSINBUV961/0AaAC9+NUtywFmwSDcdZs487HtNPW+r+/gHYZDslt0z78eMT4qTzZDV37HzMmAXrD1Q7ic18uqFM0ASow/ysWae4A3oE5ivv99WmolEWuIMP+j52MmQ/NxrS0Hkc/XADHAy/5lP++APxKO9nLguogbBmyGrCuO9JLP3guUuEY43iOfRg0hk2P8p1VJ8HYgSxgDKWyGyR6yzcBsWOvruEqCeed091qt+3Mx7NkG0xjh1psC+MMZmS1wN9/zqDM99HDM//pyAAO5wjUicpzYs2AP8BTXlVWZnwPMgxH8X5kstTGwAOIa2l0aFkPEGiJVnaWqs4CtHq+DSkeAfsAF/mXxQLuSlYEDHjZgtmUmeouPBbjEMSA+PqK6J0Ma82GZz4+xdWWUdDgPyIQ9xHnX2QzHf7WIP/K7lok6XQ9xSc50zKGEKNjgyDzuK+5CmM9xJirLoedo79EUAAnQgeWUzRd/heYNXGEJhr4ZwCV4jaYkGhgFeX/7fGCWWkvEGiIPQrYHpXkDzPJzgMSkqQ1gQVSvwD6iBkBP8HSNADQfAGyH3+iBn49oB9zCJOAbb/mF8L9AljAf1pDqHT7QBr5+oG/ZFKuMD2DR6e1p2MDDrfaHubceZUFIjlO5EB5mPOR5tO/iWg3zCAhagFkZ9KTEZQQd/08hxPUwy/9lbAbOhMnc4qU3zYF5PiuJllpLxBoiEflcRD4HTndezwh2n3GXwpNdxsBPAcqehxcYi8fv2E0vmNXrOP8Yo2uAC+Hpb2/FLAB6cDrMemUwZv+vB63hecb69/EwFBDnbQiXwWmbf+LH7Sd71y2B7ttWkPWlx1TxVOBEeJ3LykTbaQq9zDK+l4O+F/xNE4ht7pb9By7e+rZ3P0mOn4vZ5n0K0A+ezf+3u85eYA2sXtXJLUuBNwecD0/436aldhKxkdWqeoaI1FHVUgARaRz0Tvs5/pLkAGWnw7svjIRl/q6qrzP68jg3Qju8R1PbQJtAQussfOMHVtzd0rVTw4vCJMja5BcIDgkwmZth5Xa3rA0sSmlP0RneDu/stIbGb+Tp6+oFxECJx1e+gVZwIvzGMV4Gdnffuny8/BIonOkWxkCv5gv42bOj03GiwD8078+BJV2OJK9ZM3edJNhxaSws8ggFOBH+x/kmTCLQCNNS64jYEZHDfwFE5EzgnaD31hwOYxdMTvcvmwEPj70W+Nyv6LT3fjIG7COfdk62pZwn/C1bLvEcfc9CP/nmBsn0bfGjn5zZcC1PUbZCBZAJHfJX+K3yNfskjzj2wPEewq+AzdDGI1jyTzqR1SvBLOlnufe8Ndpa5DjljyiTrbuwmQlt8GQbRFOC2VEDo7o8S+fvVsM892x6x8OxNF5RSL/uHkOus+AKpsL4r/3v01IriXRD9ImI/AScBIGCa6qZYmhILkSn+5fthS2k4LsM72IYHwFHewt7gRTit28NYDOHs/Qz/2tdz1P8tPw0/wYpru0Zf7hl7SFmL/5TyUz4m6bes8E2wA74+eN+ZaJtJPEjJ5HITjyniNkpDWl4fLaXrIA4Pl1yoXc/jWEvMebCwEC+NW6lje7RT+MZhbADZm3q7243C2LYC3Tzv09LrSRiDZGIPIcZvDfCrBU9GfRO853I6uLf/Yo2jW3MhwwDAkQDH+9y5Pr4gZJgXkr3gD6nISu+4+iz/EdE0xdcRLMO6/3kT146hiuXT8MzDOCXW7vwfeM+/s71M53gS0+GARcCTd2iP+nEhbM+5db8SXiu6s2kP8XFUXgavQ20ok5T77ifhf852oy8nO0rJ/GDWXX06INoODPtfa+9ZsPHTqWYKLx3x1pqMxFriIBJmLHE2Zh95A+LSFBzXGs/8+Okof9/6r3Uc7Ze1PcrW9bqCLNEn3qHd8FsZzd/gI2yi9q3Z+mqY/zkn/YaQPZn/mv412e+iNk27x519f5wsfEFeaQAAVjRqyU/cpLJiemwKaWxGbh4LLD9TVPm9etOVHQJnmv1F675lMKZjb362kArShc38Oonjj1e4QA/k2a2Gxd6VDoaXuJfFBW6t328M3sU22gOPeyIyGKIZEP0L+AozJ/2Jsyo6O7KNhaR1s45WFNF5KHKtNkb42wsbepf9gfdnC0O/v79rTRnK81hbbp3QROYT1rA7brbaA7P+9vVDvwVcAkhu0dDzuj3AV5xnSfC95wCePe7llQvAwGwnA7kt6nj5ZSOopg/6MoFMf/DhGMbvm7T1zFYw8tkf9AVPvLW6UdOcuKDzPF03zCQdT2aea8sFmJGP2+7P4T8XnW4l7thGRYLENmG6A7MeVbTgLeB5qoaYJ2pXPrhHvv77SJ1jjJOF5F0V0BeboOG/MBJsHam38U6sNyJLN7mV7aZwwPmEMoakGB+wAF+cD3IhC6AT2zOFK6Awbv96l/Ie3yf0x+XYxiAfGcP1/HedXu6wqQ9ZmdRlPBnTCevuk3Zzul8ySvzr8EE9ria7XJ2xr9fJjub6TR+wn9auvGZ9rgCqMbyggmSzPKo4Fou9AjEbLC5lHXr2kJhut/1LLWTiDVEqlqqqq+q6jBVvUBVX6+4lRc/A6OAfwI3iohX9JyqzlTVdFVNd+X2aJaZ5xgbfx9Rxxnr+OnJ0zDOFm8uyPmYNH4GLvKSb6eJiUYOEEDcbFme82P3NmwpbIb0Rn71Zy4+g8IbGuNyDAOQ7+Q08tl2El1SYqaRHhv7+738M712LPZaLk9iK0esyebItCWY8y8N8eTSrOd6PC3ogDVz2THZe5Pqmcyg39Xu1bBU1tC0ZLv3iHKF4+T3NMZr4JUjbP57i5uINUTVwDFAPScOKY9K3OuiHu0ZyDdAgF3hR8Il179KWfCeB3V3wF8cBbzrJW/CdrPZs9CvCaM6PsvR3Rfi67C9vOR1EsZn+dX/pUsXjpjiM7TKdLZppPpf/086ee3UJwX+btzQK41JPLnQAPZRD3jOq232ktZ4Hh2mCfhNMZfTgVyPjGq7SGRnVKJXsPjtl9xFIruoO8o9ylsxoKUZ+flMKS21l0PZEK0AJonIM8AMVd1bUYNOOSt4ffvlBPpl57epQ1tW4rcyBvAVnMtH0DDdS5y8JodjSn4L6PN5bcJVZgrkQ6NPihgY842fvPeGxVzBFG9hJnTlDz8DEZ9TZOTtPIQ50Gx+nlfdaErgWxjNy3g6q5uw3W+UJZk4RsztwB4wYy67OKzs/Taac8SKbK+R2Mn8QMfF6+jaxD3KbLdjI/XYS8AdxJZaSY0xRI5BqTSq+quqXqSqV6vq8xW3gJcTRpLR5B4CGZsGs0pZSxvKdpp7MG9sd7OMneedWXFFm5bMiTqBLqf+golC8NBviPBbjn+KwoXDjvZPtwGMaPWK45j2mGFe7TiRfeIl5VtnRJTq0d8g0PZQJ9V7CX7UJc8aB71HWMAxJb/RZdAvXvWeHnAlx3WfhdcIrhgnBsnQZ3mm2XfX312lIbkUtfLeuf9D4+OcJGv2sFaLocYYIuDaYHdw1bbXuH3+4wSMFSqGaVtH4Lm65CJtxyKzDy2xjZd8K0k0YTuLJ/TGd9Pr6hOTKVzQGNeKk4tVtOXnWf3wZVrmlU5ysXFu4TZnRcp3xNXGCRj0WOWSGSa4svRT9xJ8MVFczDs8uuROPKec/4p6yew187Ao7VjJz8/0w8tZngILV52Iy+JN7TCc/CZ14GZ3lb6zF7ItoTGLf+1dJiugPrfwCLTz/rwstZeI3WsGICLvYYYACsSICKoahAhrExuUn1DHSSKfhN9/617QJGk72YzE1xckxbByezszJdnljjOKYZ+ZfkWDmSe5DVwmxxDbaweFzMYzNmkvMc4Ipy5eX08SLF10LMav4tRfgfExT8XrGnqkazNqkVvuipEc7667jxizo38xwHG4Rju5xJM1+0jg9bK6gzfMou7Fuym6ZqX7mq2B6eK0q09zttJgW6mz9O/USXF26K90y5LYxsLlfWHlHgLFZVlqH5E+IloHXOI8lgfHCIFxTrdgfkwaV3Z+GlpeWiZzPV5LusjZp5XrV3Z/0n+4rclEWKxe8kR2mcT07cCsdrnLzsz/nMK1jTGFbvmD3O4YliO85BekvOFMjRqXyaS3Ouk68Kr7ZeN+jv/pEbd8NOxIiYW+YEYwLdhFIsP4iPbDF2GMq6n7GDfRsEc2ZkRkZK+2uoSi8Y28+tJonIRqZ5fdb35SHSdEwNT5qc2xZto61S1LZCcXdXgNiCPgwoCl1hHphiheVQswySQOq6hy1UkFUklhM6+sGwsbt5fJXI/h+e/y81X9MFMz77LRvGyyLbYTnzJYRTtn6TrNq+ydBhc5q0tneMlf5F/Oj9u7j//Nv4yfN6VB9HVlMs0Vk5isi3f97znFbGQ9/o4yWVZSAn/RwQklaA6ksoFWTOEKVkzojhlpmbqJ7CJvZTNMOjgj60Gmk2DffR8vN77Uw8iaIMrfYnrAELc+93G38UEVumXN924zydN6uGSW2k6kG6IPROQj4BPgjeB10x5oT8cP13HpEVMxo472Xo/YHXD0cwsxAYjeZRtoZZKTrdztJY8n15w7lgjmB+cu68SfzuhkoZc8j3gvncoereC4FvOhWMtk2b0asoY2znTSXXc475hkbPPcdZNn53D8mkXOdpN2QHs2b00hjfn0fehrjCEydRfQk8ZdNmHW/41sFW2diOmO7n5K3nHu4USgPXuIo1f+Qmdrialjzj3DyXppZA1eLjUrkMUumaW2E9E+IowXZDrGi+G/E7TaMBkGs4c1JJMemOPUvE+seKzVVRzOZpZyGpQZC8McTjCbTBMbwS532Y+cZEYm5wLj4/H0h6SwmYQeWeTQH8+vYTYnePip3GSlJPDzon7GqO0yZX/Q1eSl3uhdv13JSqZF/QOiBYqNfN2JzWienw1vbsc1HSot3styjjK7/ZNPgyyTvmM65xAflcuOdufDSiP7kGHUvXQ3RTe0wZU0s15hEUemLWE1TYEkttOErQ1cuYhMvyOYZnSM9tDxTCcXkqfMUquJ9BFROiYwcQ/wbNB6SRZIFnJpaCKhe7hlrsdN254zo5XERn5lJiYGk3zMQ34Uy6nPHrN61TLOq2wDrciJXQbRdb3kZSex+vQxhSuMwUl2l0VRYmJ+fPq9OuoZc5JGF7f8iNnZRBUDM5tAS6furlg20NqcbNvQXTeFzWbqlOeWTeIWip5u5NV/bA6s/qwzdGkOycJvHGOmqIt/LasTn59ngi49dHm6zZVmr52rT0utJ9JHRDuAPar6h4j4hxtXFx77rxbQy5wX6BNKVBTjrGgdj18AYQFxTN96jvH5eBzO+CedGLn9dZMGpAden/YfdIV5/c00xiOusQPLA540m8Z8s3Ui1d3/r/Q09X3Ckd7IHMObPc43vhpXipA2MDhhuvFX9XLXHc1LPJR8r5lxOfqdyjekv/KwqefI/qaJqeMh+y6ljzk4IBVoZyK0O7AcLuhZlnFgc4Nk85kuo+xz7sFvzKQ//LQRzvU9ptZSG4l0Q7QdGCwiU8Ejcq66STVPiexi1pLB5keU6l2lbj5saNjK+xQNhxFMM4nwJ+K14TOGfdzd5F7u6jvZjGY8Pu0eZBoHbrR3X03Y7icDOCF/LrEdd1CY2rjMEB3OZpPHOs+nfgMnjmijh3wBjD/nYb5LHFIma9xxEzHso/HFm9gxswWu3RolRJt9rB6R2QXEmfAED1ku8WZ62c5EVP5JJxM2MIqybR5Hbshic6uVXjquoh31KYD7W+Kb999SO4l0Q7QMeA+oo6p/VlS5yjipVqdzDpd0fpW3zv2n36jntpR0UlnDjov9l5s/5FyTz/n4fl7tnuI6EtlJs9vWkz3fO8fQOXxM+36LWBHb3Us+hxNodqV//WkNLiGupIDCSyn71uLJ5b9cZ0ZQ7i1fLGt/BF8yyBgER59laUewllSOHL6E1anmRNoToubwIcPYkdXE6+igKYwyBnegW/Y6l3t9VmBSnOQsSy6TbaAVG2hFwsAscpoa4/RBqzPYShJHXr+E1fNNv+8w3EzXhhBwH56l9hHphmgg5oTXHSKSpKr+OTjKQURaAI9ipndLVLXcLSI3pJl0RSMz32V5jw7c0M8/fdEpfE88uZyc5p9P+qrFr7G2SxtO6OedbeTxT25n0TnteZnRRKeVeJXdxb3cyQOcleZ9OMmd2x5la1KSX/0eZHJ21Cc0PXG7+REDQ776ju2nN6XHqZlO6lVDxw3rGNFqGvX77XFySkNq/jpSYjezMqod+9JMfur6FHAmnzG/RRrNW2wtu+65fMQxd2R6Hdz4+ILbKTkxysnIaHhw9n3EnLi3LBvkQ/m3s7xBe66ImVK2mf+8xZ/zdZe+DOcdCtKMs/7xFbeT0f5WOnV3/W+Z4PeZWmoXohqyY8MOGBHx3OWpqvrPA2h7L/CFqs5xjiIaqqoBNzeli0Tuh1ALSFetksc6PT1d09PTq1kbS2UQkQyTQqd6iNgRkYgMVtUrRCRaVQMkW62QZIwrFYx/KQEPj4SI9MfZTPX999/Tv3//g9LXUmXsspklcg0RMAJz7N+LmORmB8p6jHt1AyZC0SvnhqrOxH026z1VVdJisRw8kWyIXFT1P+bLwGMiMgr4oIqjKovFEgIi2RClishFQBvnGVV9t4I2ZahqFmZUZTkQSkuhTqTHuVoONSL5L+4VzJ6IKc5zePNFFO8nwWMEO/wBo19JAD99foAgnnuDuLd4bx4s9z8p12KJWEOkqq/5PoLe6Re3Q245Adz3l7Mnav08yAiQHR9g4wIoKWdGWJjjL1s3F14/11++L99fBvDDJH/Z3jzYvspbtvh9uK8p/PZ/3vJJbQNft6jA+/39yf73UVribYALc+DbeyHHI/1reoJ3n3lb4W07SLX4E7GGKCzMewZ+mFx+eVGA6LvdTrKzOf/1L3v5VHgvwGkVs5+Cif6HKDJlMKz+3l/+YAq8/Q//kdd39/uP1J7rA/891ruu6/VvARIYzApgzH58zPt9cQHc18Rbdm9jbwO8bg78+Chs+Nm73sf/hq1OvFCdaNBSb2NlsWANkT+/vFR+2RNdAwgdX/pXdwZus+xTfwPy9V0HrteyT80Iy5eXT/V+v8tJUvDdfW6Zlprnnev8239/v7/sh0f2r0vxPn/Zdw84bQMYNpeRjnLybT/eef/Xt9Q6rCEC88Py/HHlZrllnmX528zUp6jQPIr3QolHuz07/NsAfDHB1PWVpye4r+M5sln0TuDrvDLQTJs85Vl/wN5c/7o/PmqmdMV7ociJhs7dbKZQvnWnnOnWz8Xm3/xlBTudth5TN9c9u4zdtj+de/Jot2ga7NvjrgOw4uvABs1SK4nkVbPQMecp7/ePdoAB5YxaJrWFfrfhysnDPI8DQh5pAwPuxC/iYP5zUD/RTE18+ewmaOLjq/lwDORs8K8L8GQPOO5fUOrhs3moZWB9H0yBU++GhR5TsomtTV3PUdq62TDrYagb55a92B/6jXePYgAeTjVtc7f43PNdsG2JW/Z0b+jpMyV98HDocan7/f9dYNqdfDMWizVE4P4xuKYz6QEcya6yO70PRKSo0D2V8W3nanPXdoiKrlj+7b37rz8wHfreaF4X74OZD/n366rb+Ty40NkhU78xfHqDd93SUve07K6/3QYn0GfgK9u2DH552Vu2YArsdnw/N/7h3S6uCdy6Grb8DplvGtndO6BOFBYL2KnZgXHyrf6y+GR/mS9R5dj7A5WnnuR+7RpdHXlK4Lote3u8CRBeIB6jNs9RT2UoDbASWOw48ntd6V929NlOnx5/btYIWTywhsiXtgPKL+sfYJd4bIJ5vqiclNr14v1lcU38ZRXRrCO09Mho5go6HPRg4Pp9/u0v8zROLkM0NEDiyxOu278uSZ38Zac6U8NABqbvTU6fjs4DynHsW2otdmrmSesT4B/vl18eKOL4yFMgbRx0Oidwm15X+MtuXb0fHfoEEApcNTdw/eY+RiGpEyQeEbjuaJ+jrBOPgGP+4V/v9Pv8ZTd6+IDq1IHhb0L7092yrhfB9OvNlNCT816Cwxx9XIbo5FsC62eptVhD5Mk/9xP123FIYHmDJnDGw4HL0nMOLOo6qh5c9pG/fPz6ym+7OP0+aHW8tyypszEIvtzwu7+sRwDDlNAaEnxSurqmWy7qxfn7yG5bC/U9IrXjmhgHuMXiQ0TnIwohkfEhLP4AOg2tvP+kqBDqxlZcL7KpchoQm48ofNSafES1ki7nVVzHk5pvhCwWwI6IABCRacBfAYpSMacMBqK8sgOVgznfYl4N6r869TpKVau0AU1EXsY51c0Sclqq6uiKq1USVa31D2BmOfL0/bQJWHag8prYfzXrFbBv+6hdD7t8b1hbjnzmftqUV3ag8prY//6udaBtyuvbUouwhsiwNpBQTTrZgJRXdqDymth/depVXt+W2oU1RIaZtv9a2bclQrDOaovFEnbsiMhisYQda4gsFkvYqXWGSERaiMjbIvKsiFztIR8oIq+LyBsickIY+h8pIl+IyFQRCZBftlp1aCciv/nIQnL/++k/ZPdviTxqY2T1WOApdY6iFpEX1RxFfRNwLhCiH6WqAAAgAElEQVQFvA0MDXH/JwObMN9JoODCakFEkoHRgG9G/pDc/376D8n9WyKT2miIyjuKWlR1H4CIBHPvRHn9vwr84rx/mSAZAjXnvY0XkS98ikJy//vpPyT3b4lMat3UDPdR1OB9FHWhiNQTkfpAgOM6gt7/iUAxsJvwnAcfqvsvj3DfvyWM1Lrle2dq8BiQCywAugH/AfoAVwJ1gUdVNcCRGUHt/x+Y6Ukd4HlVLScBUbXp8YWqDhaR/xLC+99P/yG9f0tkUesMkcViiTxq49TMYrFEGNYQWSyWsGMNkcViCTvWEIUAEQlw0H3Q+qojIi1C1Z/FUh1YQ1SNiEhTJzL5JSd62nUQ2QtO+eQqXNM33qYiTgIuE5FkEanyMaoikiIi/wkgTxWRG6p6XYslELUxoDGYHIPJr3M3UA84R0QGAkeJSB+gi4ikAu8CM5z6C4C2wIfAYUChqr7tWt4G8+MHxqvqOBEZj4k87gy0Bw4H7lDVlY4OpwEnYNJrtBORUcCZwCrgSCATs1R/HdAOGIaJpl6gqi973Es6cKuj978w/7Q+VtUPHSOVqqprq+NDs1jsiKgaUdWvMT/0x4GngJ2q+g2wwicuZqWaExDygDeA+4FBB9hdqtN+CuBxGD3fAF8BWZ4yVZ0AtFLVh4BPgOOACZh4pl2A75GxbVR1F9AciHHa/OyUzQcGH6C+Fku52BFRNSIilwHLVPUGEYkGvsQYBl92O89FmCjmWMw/hWJMQCGA53GwgeT/wxiRKzAjm6cdeaDAsN0+z8VOf9EYI1gAjPJps9d5Xg7ci4l8ngxc4uhdGqAfi6VKWENUvXwLPCUihZip2ZuOPFZEzqxE+x+Bl0SkF27DA7AZaCwiT2OmV59hplfdMdMqz+NpN2GmYu9Vor8HgakYwzTdp2yDiCQCzYAbMKOuH5yyvrgNn8Vy0NjIaktARKQJMFJVH/ORtwFOV9UXwqOZ5VAkpIZIRP6J2VdUgtlxPlJVN+y/lcViOdQJmSESke4Yf8mxqrrBWQLuqKrjQqKAxWKJWEJpiG7CGKFLQ9KhxWKpMYRy+b4YjxUdEakvIh1D2L/FYolQQmmIvgcGisjhzvuxwCPlVRaRWBG5XUS+EpHpInKLiESFRFOLxRJSQu2svhS4xXm7Bfinqm4up+79wLuq+rvzvgcwTFXvCYmyFoslZET08r2IJDs5jhGRzqq6JNw6WSyW6ifSt3j8KiKnOq+vDKsmFoslaER6ZPWbmI2jLQi8dcFisRwCRPqIqFRVrwfaEMTjZaZOnaoYQ2cfh/bDEqFE+ojoaQBVzRCROcHqZO3atcG6tMViqQQRa4hE5HNARaQpZjsIwNf7qX810BuzWbQv8A4mfUUCMFZV95bX1mKxhJeInZqp6hmqeiawRFXPdF7vr/4zqjoK2IhJ9pXgTOtmAecHXWGLxVJlInZE5EGl5/ZOpHYCJr/PRke8EZMuw7duf6A/wMiRIw9Wx5pJekK4NQgt6Tnh1sBSDhFriEQkCXP0cH3nNaq6rYJm/wYmAfsAVwL5lph8Pl6o6kxMOlXS09NrZ5Ck/WFaIoSINUTAw85zofNagX9W0Kadqq4CEJEdIvIkZoT0r6BpabFYDppINkRPqOoiT4GIdPeVeeLpR1LV24OpnKV2kZmZeZuIjMOM0g8lVFWf79Gjx8MVVw0ekWyIujlf/CJMjuRemHxG5RoiiyVYiMi4zp0774qOji4Jty7VSXFxcdSSJUvG4p6BhIVIXjV7A3PKxBYgG7hNVd/ff6vwUVJSwtatW8st37lzJwUFBSHUyFLNSCQYoby8PMnOzq4wC8WaNWvqVlQHwLmnsNuBsCuwP1R1l6p+rKqfqOruiluEj59++omvvy43zImnnnqK7du3h1Ajy6HIlClTGi9cuLB+RfUmTJiQEgp9qotInprVCJ588klWrVrFSy+9xNChQ+ncuTPPPvss2dnZPPjgg9x3333ExcWxevVqmjdvzrhxNjOuxfDVV181mDp1apPs7Oy6CQkJJfXq1SsdNGjQ7pycnKj58+c3KCoqkquuuir7448/Tty5c2dU+/bt9y5dujR26dKlsbm5uXU+/PDDxC1bttR744031o4bN65Vhw4dCleuXBl73XXXbVu0aFGD9evXR7du3bo43PdZGSJ6RCQiF/k+wq2TL5s2baJNmzZMmjSJIUOG0LRpU4YPH07btm355ZdfUFVeeeUV+vXrx5AhQ8KtriWCeP3115u8+eab6/v375977rnn7po2bdr66dOnJwJccMEFu+66666sr776qtHmzZvrdu3ateDcc8/NSUtLyx86dGhO27Zt91144YU7ExISijMzM2NFhMmTJ29p1arVvrZt2+7r1q1bfk0xQhDhhghzmF88EIdZgo8NRidvvvkmc+eag1i/+uorRo0axbJlyyrV9h//+Ae9e/fm/feN+2rq1KlkZWXRs2dPVJWGDRsCIFK5xZa5c+cyePDgMn2ee+45brzxRjZu3FhBy8rj28eWLVsYNmwYP/zwAweSn8r3OitWrOC8885j/vz55V7Ht81jjz3GbbfdRlZWVrl1Jk6cyJ133sm2bRWFkdUsioqKBGDHjh1RO3fujALjawSIi4srrVevnpaUlMh1112X3aZNm3133HFHiuvv6IknnmhWUFBQp2vXrgWqKvXr1y8FqFu3rhYX1xj7U0akJ0Z70LUMLyKPqup/gtSP14cQExNDkyZN+Pbbb+nYcf9ptR955BHWrl3Lli1bKCgoYPjw4cyZM4e9e/fSs2dP/vjjD15++WWmTJnC+vXrueee/cdO9uvXjx9++MFLFh8fT1JSEgsXLqRRo0ZVvEs3gwcP5ssvv/SS1a9fn/j4eJ599lnOP79yO2IGDRrEV1995SWLioqiadOm/N///R+nnnqqX5u+ffsye/ZsL1mjRo04/PDDWbRoETExMZxwwgllRsizTsuWLfn999+JiqpyxuAqL70vWrRobffu3f8GSB3/Wc8Dbb924lm/+so++uij+Lfeeqtxfn5+VGxsbGmDBg1KTjvttNwdO3ZEdejQYW9qauq+V155pUlOTk5UQUFBnVatWu077rjj8qdNm9a4devW+7Kzs6M3btxYb9SoUdunT5+e8M4776y76aabUq6++ursyy67rM1LL720rnPnzhXusVy0aFHT7t27px7oPVUnkW6IHsaM2uoAxap6WzD6adeunb7xxhv06dOH5cuXs27dOgYOHEidOqEfMM6dO5eMjAzuuece+vTpw9y5c4mOjqZXr16VHlUdaB+5ublMnz6d888/n5iYmCpfZ/v27Xz33XcMGzaM6OjA7kffNrNmzeKwww6jW7du5db55ptvSElJoVOnTgd76xFliCIFa4gqgYicBGSrauXmSlUgPT1d09PTg3V5S+RgDVEAIsEQRbSPSEQeAa4DeonI5HDrUxkqiieyWA6EysYD1XQiffm+PrBCVd8Ukb7hVqYiJk2axHHHHceGDRto2rQpLVq0oGvXruFWyxLB3HTTTSmrV6+OiYuLK9m5c2f0u+++uzY+Pr7UVT5hwoSUjIyMLcOGDWt79NFHFwBMnTp1vWedQ4FIN0TRwGFOyo6kcCmRnp5OYWEhxcXFJCQkkJ2dDcCdd97JzTffTElJCenp6SxfvpzS0lJ+++03Bg8eTMOGDbn77rvJzc0lMTGRkSNHMmbMGHr37k1KSgqJiYllTtmnn346XLdnOUCqe5r16KOPbmzbtm3R3Xff3fzhhx9utnHjxnq7d++Ouuaaa7IXLVrUYOPGjXWHDBmy65FHHtkC8Omnn8ZPmjSp+WWXXbZ94cKFcbm5uVFHHXVU4ahRo3bcdNNNLePj40sSExNLHnvssYBHdUUikW6IHsKk9jgDuKG8SiJyAnA9kAesAxpSzdkZr7nmGvbs2UOHDh1QVSZPnkx2djZ79uzhsssuo0ULk3UkLS2NFi1aUFxcjKqSm5vL448/zrXXXkteXh69e/fmgQce4IorrqBz5840b96cQYMGHax6lhrMtdde22rr1q11e/bsmX/xxRfv3LJlS/TEiRNTBgwYkN+tW7f8li1bFn366aeJq1atiunSpUtB796995x11lk53bt3L5g6dWrTbt267Vm0aFHcG2+8obfeemtWfHx86Ysvvtg03Pd1IES0jwi4U1XHq+ptqrp+P/UOA0ar6pWYNLHVnp2xtLSUwsJCr/c7duzg7rvvZunSpbz99tuAd7xQaal79KyqqCr165vo/KioKE4//XTOOussHnjgAfLy8qpDTUsN5L///e+G1157bW1WVlbdiRMnJjdr1qy4ZcuWXv88hwwZsuv9999fm5GRsRWgYcOGJcXFxdKnT5+8Z599dtOgQYNyAIqLiyU6OjqyV6ACEOkjon5O0vxd4J3mwxNV/UwMdwCzAZcFCJidsSrce++9gJmm3XjjjdStW5d27dpxzTXXkJiYyGmnncacOXNITk7moYce4uKLLyYqKoqGDRty880306JFC+Lj472umZmZyQ8//EBycjINGjSoDjUtNZQuXbrsPe644/Lvv//+lq1bt967YcOGmO3bt0dt2LAhpri4OOBqX1paWsETTzyRdPHFFx/RrVu3PaNGjdp57bXXtqxXr54efvjhRaG+h4MhYpfvRSROVfdUsm488AQwDfgLuENVx4nIFcBeVZ22v/YVLd+np6czevRoWrZsWWn9LRFJtSzfRzrr16+PfuKJJ5Iq6yOKhOX7SB4RPQ9cLiJjVfWFCuo+CbQHrgBKgE3VmZ2xRscYqULBTsjdAru3mOfcLEgbA7G1LGd1LaF169bFNclRDZFtiFwcD+zXEKlqRSlk/TgkkucXFcDuzcaw5G7xMTYeRic6BuIPdz8aHQ6lYU+tY7GUURMMUVDmjhGdPL+kGPKzIXezj2HJ8jA8m6GoEOKT3cYlPsW8b3GsY3Scsnpx4b4jSwjJy8uTgoKCOs2aNasx/20i2RCliMgMoJ7rsMWKzjYTkXbAe6p6jIg8jbm/5sB/VHV18FWuAFUo3OU/atm9xW1ccrOMEarf2DEuLoOSAq3SoFGKe2QT1xiqaf+ZJfIZMWJE62nTpnmtHg8fPvyId955Z52nbMqUKY2POuqovYMGDQq4FBvoOuEmYg2Rqg4UkTRVnQ8gImfsr76IJAOjgXwRaQh8oaqfisj5wGn4TO88p2YJJ17C1MLPDkrfGPaRJDtJZifNZSfNZQfJ4nq9k+aOvIgotulhZOlhbKUxW/UwtuphZGl7tulxZGljskmgeE80BEzouA8TKrUuUKFlP6ydeFa4VfDjpptuStm3b5/89ddfsccee2x+ZmZmXGpq6r6ioiJJTEwsvu+++7KuvPLK1nFxcaWlpaXi2l2/dOnS2PXr19cFyMnJqTN69OjWpaWlDBgwYPe8efMaLF26NFZE9I033miSn59f59FHH900bdq0w9atW1fv559/jq9Ir1ATkYZIRFphDEd7EVmBWe2IAT4vr42qZgHjReQLVc0DPnVGSMOBKwPUn4nH1Cw9vZw/0tISM0LxnBLlZvmPavblQ8NkZxTjjGDij/KYNplH/ZiGNALaHcTnYzm0GDNmzN9vvvlm40suuWRnUVGR/P7773Fff/31qpEjR7b64IMPEs4+++xdw4YN233llVe2Lu8al1122fbVq1fXmzVrVnzfvn3zOnTosPe5555r5lrGnzt3btySJUtip02btv68886LuN99xCkEoKobgDNF5FRV/bYq1xCRc4FTgVEVhgHsWAOZ07wdv7s9pkmxCd4+mEYp0Oo497SpUYqZSoUhbYil5tOgQYPS6OhojY2N9fKHqqo4z9SpUwcR0aioKC0uLpacnJyyP7a5c+fGTZ8+PWHs2LF/z5kzp6ErqLa0tFQmT568ecmSJTGegbaRGPAYkYbIgy4iMh4n/kNVB1amkTMSegn4EnhRRN5T1Y996vTHmZrdOqgVrHZWlpq0gzYnuY1Mw2SIrleNt2Sp8aQnHHAaENJzKrU/zWWUxowZ07JFixb7zjvvvJzRo0e3+u677xoBDBw4MPfaa69tlZCQUHLKKafsBmjcuHHJ+vXrY5544okkVZXWrVvve/XVV5uMGTMme/jw4alFRUXy2muvrevWrVvBmDFjWv7xxx8RFz0bsQGNACIyDXOW2XzgalX9dzD6sfmIag3VE9AYREMUDmxAY8XsAZYDnXBv26gUItIa+ATIBLao6oTy6mZkZMzIyMj4pYo6pgJrq9j2YNuHq204+z6YthNVdXwV21qCSKQbok+BXOAizL6xA6Ef4MpQNqeCur+oavoBXh8AEUmvatuDbR+utuHs+2DbVqVdOFmzZk3dNm3a1Kh9Y1UhYg2RiEQBK1V1MXChiFx8gJf4GTOt2wp846ymlfeFzqy6pgdNuPoOV7/h7Dtc/ZaLTYxmiFhDBLwB7BCRdUAKEAW8fQDtjwHmqmqpiOSxn5QnzlJ+WAhX3/aeD4Jq9vfYxGiRnY8oT1WvAUYAi1X1ugNsvwKYJCLPADOqIzlaOcwM0nVt35HVb9C49tprW/Xs2bNDVlZW3QEDBuSddtppu1esWFHfNzHa+eefn3rPPfc0B3AlRvv999/j4uPjS5zEaIfdeuutWbfeemuNS5oeyYbIdUrcn6r6yoE2VtVfVfUiVb1aVZ+vZt08+5kZrGvbviOn32BiE6NFtiFq5xwx3SJSj5u2WKoLV2K0H374IeHdd989rLKJ0davX1/v4osvPiI7Ozt61KhROydPntz8rrvuSgm1/gdLxMYRiYhfbg5VfS0culgsNjFacIlYZ3WkGx0RuRroDdQF+qrqESHsu9IxUkHouz0wEcjChD1MDVXfTv9lGRZC2S+gxcXFUdHR0RGfWuNAEqMVFxdHcYAxesEgYg1RpKOqz0DZsdjDQtz9gcRIVTcJwM2YuK73gamh6tgzw0Ko+nShqs8vWbJkLJHtzqgKpZXIgBp0InZqVhMQkY7ADao6LsT9dgB248RIAYP2EyMVjP5bAFOAmar6YKj69ej/C1UdHOp+LcHjULPuoebfwKQw9HsMUE9VSzFnuYXsexSRHkChqp4O9BSRw0LVt+XQxU7NDo52qroqDP26YqSyCW6MVCDqAS+IyHZgtaruDGHflkMUOzWzWCxhx07Nwoiz+mWx1HqsIapGRGSUiPwoIm+KyDvOEj8iMrmcJkFZrRCR70Xk1mBcu4J+U0Wk0lHsIvIfx/FdXnldEZkoIvbv9BDH+oiqn2dU9W0AEXnZiXvpIiLRwMuYlPhxzuujRORMTI6dLpjNvf8HZAO3AT9hjsy+BLgeaA20Aq4DzgWOAhKBZz0OGTjPqZcpIn8BPwD3Ypzq2UCBqt4iIsuAt4DjgHlAM2C5R1iCl76qepWILMVsRu4J3OLoNsS5n/nAR07bwcBQ4BrgEczfWSJwk6pud+ocBSSq6iYR+Qn4DjgW+NWpX0dVJ4jIXOf+/+8gvhNLhGP/0wSXX4Gjndd1MUZkBfCKqv4KrFDVGU69dzAGwXVk0gJVfQCzKtYSOF5Vb8AYIQHGAQXADswpJQCo6gfOdb8CdqjqaOBqTFKwa4AYEenslGVggiJ/BG4CPE8Q8NLXkf3tLNd/DPQFVmPiiL4FznbqnAjcjVlRPB2T1K4AcwLv8R7XPxOY67yuo6p3O9ddo6p3YIwdGGN8NpZDGmuIgksfYKnzOhoTCLgSmCwiSR71HgQaYXIoufYVuYL29jllLnkjoAnGkIwHHscYskDscp7r4I6eLXWutdt5X4RZji/G++8hkL6eOtUB7sKM4uZ56LcKY9jOdOr84ug51SlzoU7f+OrioadLFvbIX0twsVOz6udqERkCxAKzVXWlc4JCETABE4S4BtgJ5InI5Zgp06mYI5MaBbhmLmaq9V/gcMyo6AsReRVoCtxZgU5PAw+KyBYgX1UXS8UHMwbS15dNmCjv/rgNURZmRDQDk8LlYicVS3PgCo+2nwEjga8r0ONUzEjJcghjl+8tYUNE/g18oqoB0wCLSF3gblW9K7SaWUKNNUQWiyXsWB+RxWIJO9YQWSyWsGMNkcViCTvWEFkslrBjDZHFYgk71hBZLJawYw2RxWIJO9YQWSyWsGMNkcViCTvWEFkslrBjDZHFYgk71hAFQEQuEJGZ4dbDUjEi8pyIrBGRB8Kti6Xq2DQglprOWKB1eTv4LTUDOyJyEJF7RWSViPxM6E9urRWIyNkiMl9EfhOR2SLS5yCv9yMmD9LnInJS9Whp8URExovIChFZKCJPiMjaYPRjR0SAiAwFzgd6YNKafhRejQ49RKQ9JhNlf1Xd7qSr/UZE2qlqlY6QVtWTRESBU1T17+rU1wIiMggYBfQGcjA5zIOCHREZBgIfqGqukzL11XArdAhyGia75LcikolJhl8KtAurVpb9cSbwnqruUpO47JlgdWRHRG48c6cWh02LQ5co4FtVHe4SiEgrYHP4VLJUQDHev4uSYHVkR0SGz4ELRSTROUPrsnArdAjyLXC6iHQEcI5R+h2oH1atLPvjM+B8EUlw3l+JOfSg2rEjIkBVZ4hIV2ABJkn8Isw5X5ZqQlX/FJExwNtiMvcXA+eoal6YVbOUg6p+JyIvAXNFZA+wBNgTjL5szmqLxRIQEekFnKCqTznvbwLSPKfX1daXNUQWiyUQItIIc7jm0Zgp2XpgjKpuqva+rCGyWCzhxjqrLRZL2LGGyGKxhB1riCwWS9ixhgiYOnWqYpxx9lH1R1U/3ypjv7ea+b0FwhoiYO3ateFW4ZAmWJ+v/d6CSyg/X2uILBZL2LGGyGKxhB1riCwWS9ixhshisYQda4hqG+lmI3VBQQHbtm3jvPPOo7CwMMxKWaqbmvb9WkNUS3nvvfcYOXIkmZmZDB06lKysrHCrZKlGatr3G9Y0ICLyOXAkJjlWCdBcVbtW4TonANcDecA6oCEQAyQAY1V1b7UpfYgwYsQIPvjgA4499liGDRtGcnJyuFWyVCM17fsNqyFS1TNE5DFVvQlARB6t4qUOA0araq6IfAWsUdWxInIFJhf1tGpS+ZAhKiqKCRMm0K1bN5YvXx5udSzVTE37fiMhMVpzEenivG5alQuo6mdiuAOYjcmFDLAR6B6ojYj0B/oDjBw5sird1mhEhLS0NAB69OgRjOv3Jwifb23/3ipLVb/fcH2+keAjGg+MAC4GJlTlAiISjzlhYB4m8X0Lp6gl5eREVtWZqpququmpqalV6dayH4L1+drvLbiE6/ONhBFRN4xPZw7QgaolU38SaA9cgfE1bRKRJzE+on9Vk54WiyVIRIIhOg/YiknUPQX4/kAvoKr/rG6lLBZL6IgEQ5QH4DiavVa3RORKzHz1O6BQVd8KvXoWiyXYRIKPaA4wyDnq+Qufsl7ABuA14OxQK1abSB3/GanjPwu3GpZaSiSMiD7FxPzsAz70KXMd6BYL1A2lUoc6LqOzduJZYdbEYomMEdFDmHO1BXjcp+xD4DiM/+jFEOtVq7EjpBpMekLFdSKMSBgRrVHVjwFE5BifMgWuw0Rf22OgLZZDlEgwRP8QkaMxZ6MfLyINVfXfTtloYBSwBnO+0gGvqFkslsgnEgzR+UA9jD+oxKcsT1X3OUcUF4RcM0u1UlBQQG5uLuPGjWPatGnExsaGW6VDjtTxn7G2Bn6skeAjOhe4H5gEDFLVdR5lb4nIR8B04P1wKFfrSU+oNp9DTdsRbgkdkWCI6qrqcFU9H+joU7YGY4Deo4r70Go9EeS4HDFiBDExMRx77LFcfvnlEb8j3BI6ImFqdpSTxqMO0FJEjlPVn52yycBHQORndrJUSE3bEW4JHZFgiDYBA53XS4DBgMsQLVDVN8OilaXaCfaOf0vNJRIM0UfA7Zg4ogxVXeJRdrmInAHsAVRVz9zfhUSkHfCeqh4jIq9iVuIUeE5V5wdHfYvFcrBEgo/oVmAcMBa40bNAVTupaj9VPQMYsr+LiEgyZrk/3xF1dV7nYUZaljAzbZrNT2cJTCQYojxV3amqO/FZoheRR0TkVxGZjXu6FhBVzVLV8UCes9x/ixOP9AUmKNILEekvIukikl4bTgwtN1I6SM7sQJ9vdRii2va9hZpwfb6RYIj+EpH3ReQ9TL5pTxoDj2JGOvMO4JrxQCfn9XZMnJIXNsFWcLGJ0XyIoNXL/VGbE6P9BDyL8ef09Smri9l93xdoUNkLqupuEenoJEdLxEz/LBZLhBI2QyQirYAXMJkVV2Cc1TF4pwJ5HtgGXAgsrsx1VXWw8+w3HbNUgvQESM8JtxaWWkbYpmaqusFZBRvnPJ+lqgMARKSuiCQBJwK5wAOY/NMWi+UAqCkZFCJhanayiGwBXhSRX1T1Rsx0aiJwLMbXo8BfYdTRYrEEkUgwRI2AnhjDMxBAVbOBf4rIYZjl99OBpWHT0GKxBJVIWDVrgjlOaB1whE/Zc8AJGB9RemjVslQnqeM/47tl28KthiVCiQRDdD3myJ8sYIxP2UagD/AU7kBFi8VSATXFN+Qi7FMzJ5BxZznFeZizzr7HpJO1WCyHIGE3ROUhIh2Bz523CnwQRnUsFksQCbshEpGhwC+YxGg/qOoLTlEfoJVP9f1u87BYLDWTsBsiYCgmmPEbzHK9iyRMHiJx3muI9bJYLCEiEpzVislbPRtI8ZDPwxywmAn8CnQPvWoWiyUURMKI6CXM6MfrXDNVnSUi56nqlwAist9cRBZvyg5QrKZE6vZARkswCashEpHnMCMiAc5wXv/kUWWLiLzjyCtMbOaTGO0WIBVIAG50giQtYaRBp37hVqF2UgP2D4Z7RDTReQ7o/1HViSJS37zU/eat9kyMJiKxwMmqeraInIKJU3qwGvW2VIEGnfqHW4XaRXoCUDOS0YXVEKnqOhG5HpN9UZzHqT51KnWemapmAeNF5AtMHiNXGO9GvH1PgEkABfQHGDlyZNVuwEn+Tl0AAApKSURBVFIufp9vcu/gXNdSrYTr840EZ3Ua8BYmwjrg0Q5OxsUDYRtm6wiYXfubfSvU2ARbNQSbGC2yqGykdbg+30gwRHswBqgTUOpZICIZItIJ+FFEnqjsBVW1GPheRJ7FbBt5phr1tVgs1Uy4fUQAn2JyDl2EmUZ54rczvyI8EqM9WY06WiyWIBLWEZHjSK6HOeH1f8Bcnyr725lvsVgOEcI9IooBjgS2lFN+HSZX9V7MypfFYjkECfeq2RdO0OJE3zIRmQR0wR1npIANarRYDkHCPSIC6CciczBpPspOc1XVW5zVshaYEz4sFsshSiQYoqGqugxARHyDTd4GdgPFmBHRv0Osm8ViCQHhPE6oJXAH0EtEfsFMv7rifbbZDlW9Khz6WSyW0BE2Q6SqG0XkIeA0TAoQxSNTo4jcBnQVkQeBXU6bR8Khq8VSmykoKCA3N5dx48Yxbdo0YmOraSe1B+F2Vq8HXimneB4mBYhrj1nNOLPXYjnEeO+993jrrbdYvnw5Q4cO5bXXXiM5Obla+wh3HNFRIuKng3O4YjJmyb658/rqEKtnsdQYgpksf8SIEcTExHDsscdy+eWXV7sRgvA7qycCO0XElR4WVf0ZMxWLAQqA+php2wNh0dBiqeVERUUxYcIEunXrxvLlAbeDHjThNkQfYlJ3DMYdK/Szqu4DXnceFoulHFLHfxb0ZHUiQlpaGgA9evQISh9hnZqp6hvApcDfQDYeGRoPFhFpLSKZIjLVcYpbLLWb9ISIPe8sEnbfP4hJ07EJc5JHGSIy2OP1pQd43X7AVuf1nINR0OJBeoKTcMtiqT4iwRDtUNWPVPVjzC58RKSViMwAnhaRGSLyOXDOAV73Z2AU8E/gRhGp61koIv1FJF1E0teuXXvQN2HxJlifb43/3qrJiAdrZBOuzzcSDFGciFwvItcCiQCqusHZ6nGRqp6pqmcAww/wuscA9VS1FHNirNe92gRbwcUmRquZ1ObEaOMwidFWOa89uUREfhWR2cCCA7zuCmCSiDwDzFDVvQevqqW6mDatZuRStoTmuwq7IVLVElX9QlVnqGqJT/FhwKOYlTXfXEUVXfdXVb1IVa9W1eerS9+Ip4b4b6whqjnUCkNUAXWBDZj9Zw3CrIvFYgkSEWGIRKSziDwtIuf6FD2PWVHrCiwOvWYWiyUUhHuLR3vn5UjgP8CJnuWqOldVV6nqdar6aMgVtASH9AT468twa1F7icDpe7hHRH2cc81mYU7aWB9mfWosqeM/i9hgNUvVcX2n4fpuU8d/xnfLtlVc8SAJd2T165jkZycAb6nqf8Opj8USbioyOIfqP5twT82ew8QK3QHE2q0YladsBBSBw2yL5UAJ99QsD4gHUNXPgPvDq44lVIzoWrfiSrWYQ3XkUx7hNkQ/YQ5YBEBV88OoiyWEHPKGyI5UD4hw+4g+VtXfw6lDjcP+gdcsDoHvq0GnfkHvI9wjosgmjDvNfVfB7KpYhFFdfxc1wFA16NQ/6H1YQ2SxHAwuQ+L7XNl2FuAQNkQi0kJE3haRZ0UkOPmuKxoxeZSXrXAd6B+g/YMNLRV93rX8+wjWvrND1hABY/+/vXsLlaqK4zj+/YmV9ZBGZfcwMOkiFEmh1IMgRZjR5al66AI95CUty1JIsSKzCwWVkWEhlJUPSgVaFNUhDQ21fAgsyy6kJKgZZiRq/ntYa2g8nebYnD1nncvvA4cze/ae//rPrD1/9p5ZsxfwXERMAq5pfz2ipjQoJO0LzRGNB/kf21snKigQ7fugs8GEpQcbdkVTQz/mDuaNube1JB9FREsClybpZeDRiPhZ0hJgWkTsrFs/FhibFy8F1lXY/DDgxwrjtSJm1fEGRcTM2kInr2992z9GxOIjbaTJfqtvrzdv14qY3dJvnYqIPvkHPASMybffAwZ2Y9tze3rMkvFa8fpU0V5P3663tN3MX+lZPFppEfCMpNuB5RFxsHA+XdVWOoFOtJVOoAJtPTxeK2JWHa8pfbYQRcR24JbSeVQlItpK59BIT8/vSFT9HFrxmvSGHJvRlz+sLqmtdAJHoK0ftV11ez09Xm+KCfThD6vNrPfwEZGZFedCVCFJ50paJmlB/pC8R5I0XNKXBdp9upvbm5xn+l0i6acK4lU+e3Ar9pmq+7c7+q3PflhdyGDgfmArsAxYXDSbDkg6lTQrSsuudCBpLnAacIA06cFm0mWAj5U0kPSN5i7gOODu+uWImFhVHhGxIOfzBHBDBSFbMXtwpftMV/q3ZL/5iKhCEbEe2A+sANYWTqdDEbE90gC2vS1uahlwL3BORDwObAREmpnlLNK8c690sFwpSecBgyPiiwrCNZw9uBlV7zMV9G+RfnMhqpCki4F9EXEVMErSCaVzKmhPRBwA9uXlg6Q33EDSEcB3wNPA0PplSUMrzmMS8FRFsRrOHtyMHrjPFOk3n5pV62hgoaRdwPcRsbt0Qj3QAWAW6RTnB+BXUqGoLVf9mg2PiC0VxarNHryD6mYP7i37TEv7zV/fm1lxPjUzs+JciMysOBciMyvOhcgaknR26Rys73Mh6mck3S5plaTXJS2tXUa3wejZhS3K4xNJD7QidiftDpP00v/Y/j5JZzRYf5Sk+ZL8XuoCf33fPy2IiLcAJC2SNBwY2cHo2UXACEnjSVfnGwmcDiwBdgAPkuamuwi4GZgGnE0a6DYVuB4YAQwBXoyIz3ObN+btNkraDHwKPEL6OngH8GdEzJD0NfAmcBlpsN/JwDd1I6YPyzciJkraBLwGjAJm5Nwm5OfzOfB2fuzVwHXAFOBJ0nthCDA9InblbUYAQyJim6TVwMfAJcCGvP2AiJglaU1+/ku60Cf9mqu4bQDOz7cPGy0bERuAbyNiZd5uKakgjM/br4+Ix0iD+84ERkfEPaQiJOAu4E/SmJMraw1GxPIc9wPg14i4E5gMzI+IKcAxki7M6x4mje5dBUwHrqnLvaPRvTsjYh7wDnAF8D3pZxMfAdfmbS4H5pAGO14FXJDz/AsYXRd/PLAm3x4QEXNy3B8iTZM+Kq9bXRfbmuBCZGOATfn2YaNn242WnQccT/qZg/J9td8z7c/ravcfD5xIKiQzgWdJhawjv+X/A4BD+fahHGtPXj5AGn18kMP32Y7yrc9pADCbdBS3ti6/LaTCNj5vsy7nuTivq4ncNu1zqcuzdt8hrGk+NeufJkuaAAwCPouI7yTBv0fP7gb2SrqVdMo0DjiGVGja+510qvU86YeTU4H3Jb0KnES6hngjLwDzJP0C/BERX+WcGuko3/a2kX6sOpZ/CtF20hHRStJVPG+StAA4Bbij7rErgNuADzvJYxzpSMma5JHVZg1ImgS8GxFb/2P9UcCciJjdvZn1LS5EZlacPyMys+JciMysOBciMyvOhcjMinMhMrPiXIjMrDgXIjMrzoXIzIpzITKz4lyIzKw4FyIzK86FyMyKcyEys+JciMysOBciMyvOhcjMinMhMrPiXIjMrDgXIjMrzoXIzIpzITKz4v4GGvOr0+zXRwoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 252x280.8 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "w, h = 3.5, 3.9\n",
    "fig = figure(figsize=(w, h))\n",
    "left_margin = 0.55\n",
    "bottom_margin = 0.35\n",
    "right_margin = 0.1\n",
    "top_margin = 0.1\n",
    "upper_vertical_padding = 0.45\n",
    "lower_vertical_padding = 0.65\n",
    "horizontal_padding = 0.2\n",
    "overshoot_width = 0.1\n",
    "overshoot_padding = 0.1\n",
    "frame_enum_padding = 0.02\n",
    "\n",
    "overall_ax = fig.add_axes([0, 0, 1, 1], frameon=False)\n",
    "overall_ax.axis(\"off\")\n",
    "\n",
    "hist_ax_width = (w - left_margin - right_margin - horizontal_padding) / 2 - overshoot_width - overshoot_padding\n",
    "hist_ax_height = (h - bottom_margin - top_margin - 2 * vertical_padding) * 0.5\n",
    "ax = fig.add_axes([left_margin / w, bottom_margin / h,\n",
    "                   hist_ax_width / w, hist_ax_height / h])\n",
    "overshoot_ax = fig.add_axes([(left_margin + hist_ax_width + overshoot_padding) / w, bottom_margin / h,\n",
    "                             overshoot_width / w, hist_ax_height / h])\n",
    "plot_histograms(fig, ax, overshoot_ax, \"start_distance <= 4\", overall_ax)\n",
    "ax.set_xlim(-4, 0)\n",
    "ax.set_xticks([-3, -1])\n",
    "ax.set_xticks([-2, 0], minor=True)\n",
    "label_frame(fig, overall_ax, ax, 0, 0.1, \"d\")\n",
    "label_frame(fig, overall_ax, overshoot_ax, 0, 0.1, \"e\")\n",
    "\n",
    "ax = fig.add_axes([(left_margin + hist_ax_width + overshoot_width + overshoot_padding + horizontal_padding) / w, bottom_margin / h,\n",
    "                  hist_ax_width / w, hist_ax_height / h])\n",
    "overshoot_ax = fig.add_axes([(left_margin + 2 * hist_ax_width + horizontal_padding + overshoot_width + 2 * overshoot_padding) / w, bottom_margin / h,\n",
    "                             overshoot_width / w, hist_ax_height / h])\n",
    "plot_histograms(fig, ax, overshoot_ax, \"start_distance > 4\", None)\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_yticklabels([])\n",
    "label_frame(fig, overall_ax, ax, 0, 0.1, \"f\")\n",
    "label_frame(fig, overall_ax, overshoot_ax, 0, 0.1, \"g\")\n",
    "\n",
    "xlabel_ax = fig.add_axes([left_margin / w, bottom_margin / h,\n",
    "                          (2 * hist_ax_width + 2 * overshoot_width + 2 * overshoot_padding + horizontal_padding) / w,\n",
    "                          hist_ax_height / h], frameon=False)\n",
    "xlabel_ax.set_xticks([])\n",
    "xlabel_ax.set_yticks([])\n",
    "xlabel_ax.set_xlabel(\"Distance from snake (m)\", labelpad=16)\n",
    "\n",
    "mod_ax_width = w - 1.8\n",
    "mod_ax_height = (h - bottom_margin - top_margin - lower_vertical_padding - upper_vertical_padding) * 0.2\n",
    "ax = fig.add_axes([left_margin / w, (bottom_margin + hist_ax_height + lower_vertical_padding) / h,\n",
    "                   mod_ax_width / w, mod_ax_height / h])\n",
    "plot_modulation_function(fig, ax)\n",
    "label_frame(fig, overall_ax, ax, -0.55, 0.05, \"c\")\n",
    "\n",
    "stim_ax_width = (w - left_margin - right_margin - horizontal_padding) / 2\n",
    "stim_ax_wave_height, stim_ax_spect_height = (h - bottom_margin - top_margin - lower_vertical_padding - upper_vertical_padding) * np.array([0.1, 0.2])\n",
    "wave_ax = fig.add_axes([left_margin / w,\n",
    "                   (bottom_margin + hist_ax_height + mod_ax_height + lower_vertical_padding + upper_vertical_padding) / h,\n",
    "                   stim_ax_width / w, stim_ax_wave_height / h])\n",
    "spect_ax = fig.add_axes([left_margin / w,\n",
    "                   (bottom_margin + hist_ax_height + mod_ax_height + stim_ax_wave_height + lower_vertical_padding + upper_vertical_padding) / h,\n",
    "                   stim_ax_width / w, stim_ax_spect_height / h])\n",
    "plot_acoustic_stimulus(wave_ax, spect_ax)\n",
    "label_frame(fig, overall_ax, spect_ax, -0.55, 0.05, \"a\")\n",
    "\n",
    "ax = fig.add_axes([(left_margin + stim_ax_width + horizontal_padding) / w,\n",
    "                   (bottom_margin + hist_ax_height + mod_ax_height + lower_vertical_padding + upper_vertical_padding) / h,\n",
    "                   stim_ax_width / w, (stim_ax_wave_height + stim_ax_spect_height) / h])\n",
    "ax.set_xticks([])\n",
    "ax.set_yticks([])\n",
    "visual_stimulus_bbox = ax.get_position() * array([w, h])\n",
    "label_frame(fig, overall_ax, ax, -0.1, 0.05, \"b\")\n",
    "\n",
    "fig.savefig(\"hpp_figure_foreground.pdf\", transparent=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open(\"hpp_figure.tex\", \"wt\") as fout:\n",
    "    x, y = visual_stimulus_bbox[0]\n",
    "    w, h = visual_stimulus_bbox[1] - visual_stimulus_bbox[0]\n",
    "    fout.write(r\"\"\"\n",
    "\\documentclass[border={0pt 0pt 0pt 0pt}]{standalone}\n",
    "\\usepackage{graphicx}\n",
    "\\usepackage{tikz}\n",
    "\n",
    "\\begin{document}\n",
    "\\begin{tikzpicture}\n",
    "\\node[anchor=south west,inner sep=0pt] at (%.2fin, %.2fin) {\\includegraphics[width=%.2fin,height=%.2fin]{visual_stimulus}};\n",
    "\\node[anchor=south west,inner sep=0pt] at (0, 0) {\\includegraphics{hpp_figure_foreground}};\n",
    "\\end{tikzpicture}\n",
    "\\end{document}\n",
    "\"\"\" % (x, y, w, h))\n",
    "\n",
    "os.system(\"xelatex hpp_figure\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.0155038759689914\n"
     ]
    }
   ],
   "source": [
    "a, b = visual_stimulus_bbox[1] - visual_stimulus_bbox[0]\n",
    "print(a / b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.6275, 3.6   ])"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visual_stimulus_bbox.mean(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.75"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.4775"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.75  , 3.4775])"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(ax.get_position() * array([w, h])).mean(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 10000)"
      ]
     },
     "execution_count": 313,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = subplots(2, 1)\n",
    "fs, sig = scipy.io.wavfile.read(\"stimulus.wav\")\n",
    "sig = sig[int(0.95 * fs):int(2.45 * fs)]\n",
    "axes[0].specgram(sig + numpy.random.normal(scale=100), NFFT=256, Fs=44100, noverlap=int(256 * 0.95), scale=\"dB\");\n",
    "t = arange(len(sig)) / fs\n",
    "#axes[0].plot(t, sig)\n",
    "axes[1].plot(t, sig)\n",
    "axes[1].set_xlim(*axes[0].get_xlim())\n",
    "ylim(1, 10000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Per-subject histograms (for supplement)\n",
    "==="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_per_subject_histogram(fig, ax, df, is_long):\n",
    "    df = df.copy()\n",
    "    df[\"response\"] = -df[\"response\"]\n",
    "    const_responses, mod_responses = [df[\"response\"][df[\"condition\"] == c] for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "    bins = concatenate([linspace(-8, 0, 40), [1 if is_long else 0.5]])\n",
    "    ax.hist([const_responses, mod_responses], bins=bins, density=True)\n",
    "    \n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "    if is_long:\n",
    "        ax.set_xlim(-8, 1.5)\n",
    "        ax.set_xticks([-7, -5, -3, -1, 0.5])\n",
    "        ax.set_xticks([-8, -6, -4, -2, 0], minor=True)\n",
    "    else:\n",
    "        ax.set_xlim(-4, 0.5)\n",
    "        ax.set_xticks([-3, -1, 0.25])\n",
    "        ax.set_xticks([-4, -2, 0], minor=True)\n",
    "    ax.xaxis.set_major_formatter(FuncFormatter(lambda x, pos: \"{:.0f}\".format(-x) if x <= 0 else \"miss\"))\n",
    "    ax.yaxis.set_major_locator(MultipleLocator(0.5))\n",
    "    ax.yaxis.set_minor_locator(MultipleLocator(0.25))\n",
    "    ax.set_axisbelow(True)\n",
    "    ax.grid(axis=\"y\", ls=\"dotted\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_per_subject_differences(fig, ax, df):\n",
    "    const_responses, mod_responses = [df[\"response\"][df[\"condition\"] == c].values for c in [\"rattle_const\", \"rattle_mod\"]]\n",
    "    difference_pairs = (const_responses[newaxis, :] - mod_responses[:, newaxis]).flatten()\n",
    "    ax.hist(difference_pairs, range=(-5, 5), bins=50, color=\"C2\", density=True)\n",
    "\n",
    "    if scipy.stats.wilcoxon(difference_pairs)[1] < 0.05:\n",
    "        ax.text(0.4 if median(const_responses) < median(mod_responses) else 0.6, 0.9, \"*\",\n",
    "                fontdict=dict(fontsize=12), transform=ax.transAxes, horizontalalignment=\"center\")\n",
    "\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "    ax.set_xlim(-5, 5)\n",
    "    ax.set_ylim(0, ax.get_ylim()[1] * 1.1)\n",
    "    ax.autoscale(False)\n",
    "    ax.vlines(0, 0, ax.get_ylim()[1], lw=0.5)\n",
    "    ax.yaxis.set_major_locator(MultipleLocator(0.5))\n",
    "    ax.yaxis.set_minor_locator(MultipleLocator(0.25))\n",
    "    ax.set_axisbelow(True)\n",
    "    ax.grid(axis=\"y\", ls=\"dotted\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 518.4x720 with 46 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = figure(figsize=(7.2, 10))\n",
    "gs = GridSpec(len(subjects), 4, width_ratios=[2, 1, 2, 1], wspace=0.3, bottom=0.04, top=0.92)\n",
    "\n",
    "for i, subject in enumerate(subjects):\n",
    "    df = contact_data[subject]\n",
    "    long_df = df[df[\"start_distance\"] > 4]\n",
    "    short_df = df[df[\"start_distance\"] <= 4]\n",
    "    \n",
    "    last = (i == len(subjects) - 1)\n",
    "\n",
    "    ax = fig.add_subplot(gs[i, 0])\n",
    "    plot_per_subject_histogram(fig, ax, long_df, True)\n",
    "    ax.set_ylabel(\"Density [1]\")\n",
    "    if i == 0:\n",
    "        ax.set_title(\"Histogram of\\nstopping distances\", fontsize=10)\n",
    "        ax.legend([\"const.\", \"mod.\"], loc=\"upper left\")\n",
    "    if not last:\n",
    "        ax.set_xticklabels([])\n",
    "    ax.text(0.05, 0.05, \"{}#{}\".format(\"Subject \" if i == 0 else \"\", i + 1),\n",
    "            horizontalalignment=\"left\", verticalalignment=\"bottom\",\n",
    "            transform=ax.transAxes)\n",
    "    \n",
    "    ax = fig.add_subplot(gs[i, 1])\n",
    "    plot_per_subject_differences(fig, ax, long_df)\n",
    "    if i == 0:\n",
    "        ax.set_title(\"Histogram of\\nconst. vs. mod.\\nstop dist. diff.\", fontsize=10)\n",
    "    if not last:\n",
    "        ax.set_xticklabels([])\n",
    "\n",
    "    ax = fig.add_subplot(gs[i, 2])\n",
    "    plot_per_subject_histogram(fig, ax, short_df, False)\n",
    "    if i == 0:\n",
    "        ax.set_title(\"Histogram of\\nstopping distances\", fontsize=10)\n",
    "    if not last:\n",
    "        ax.set_xticklabels([])\n",
    "    \n",
    "    ax = fig.add_subplot(gs[i, 3])\n",
    "    plot_per_subject_differences(fig, ax, short_df)\n",
    "    if i == 0:\n",
    "        ax.set_title(\"Histogram of\\nconst. vs. mod.\\nstop dist. diff.\", fontsize=10)\n",
    "    if not last:\n",
    "        ax.set_xticklabels([])\n",
    "\n",
    "fig.get_axes()[-4].set_xlabel(\"Distance from snake [m]\")\n",
    "fig.get_axes()[-3].set_xlabel(\"Pair diff. [m]\")\n",
    "fig.get_axes()[-2].set_xlabel(\"Distance from snake [m]\")\n",
    "fig.get_axes()[-1].set_xlabel(\"Pair diff. [m]\")\n",
    "\n",
    "ax = fig.add_subplot(matplotlib.gridspec.SubplotSpec(gs, 0, 1), frameon=False)\n",
    "ax.axis(\"off\")\n",
    "ax.set_title(\"Long trials\", fontsize=10, pad=45)\n",
    "ax = fig.add_subplot(matplotlib.gridspec.SubplotSpec(gs, 2, 3), frameon=False)\n",
    "ax.axis(\"off\")\n",
    "ax.set_title(\"Short trials\", fontsize=10, pad=45)\n",
    "\n",
    "fig.savefig(\"subjects.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.1+"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
